Error-related brain activity and error awareness in an error classification paradigm.
Di Gregorio, Francesco; Steinhauser, Marco; Maier, Martin E
2016-10-01
Error-related brain activity has been linked to error detection enabling adaptive behavioral adjustments. However, it is still unclear which role error awareness plays in this process. Here, we show that the error-related negativity (Ne/ERN), an event-related potential reflecting early error monitoring, is dissociable from the degree of error awareness. Participants responded to a target while ignoring two different incongruent distractors. After responding, they indicated whether they had committed an error, and if so, whether they had responded to one or to the other distractor. This error classification paradigm allowed distinguishing partially aware errors, (i.e., errors that were noticed but misclassified) and fully aware errors (i.e., errors that were correctly classified). The Ne/ERN was larger for partially aware errors than for fully aware errors. Whereas this speaks against the idea that the Ne/ERN foreshadows the degree of error awareness, it confirms the prediction of a computational model, which relates the Ne/ERN to post-response conflict. This model predicts that stronger distractor processing - a prerequisite of error classification in our paradigm - leads to lower post-response conflict and thus a smaller Ne/ERN. This implies that the relationship between Ne/ERN and error awareness depends on how error awareness is related to response conflict in a specific task. Our results further indicate that the Ne/ERN but not the degree of error awareness determines adaptive performance adjustments. Taken together, we conclude that the Ne/ERN is dissociable from error awareness and foreshadows adaptive performance adjustments. Our results suggest that the relationship between the Ne/ERN and error awareness is correlative and mediated by response conflict. Copyright © 2016 Elsevier Inc. All rights reserved.
Minimum Error Entropy Classification
Marques de Sá, Joaquim P; Santos, Jorge M F; Alexandre, Luís A
2013-01-01
This book explains the minimum error entropy (MEE) concept applied to data classification machines. Theoretical results on the inner workings of the MEE concept, in its application to solving a variety of classification problems, are presented in the wider realm of risk functionals. Researchers and practitioners also find in the book a detailed presentation of practical data classifiers using MEE. These include multi‐layer perceptrons, recurrent neural networks, complexvalued neural networks, modular neural networks, and decision trees. A clustering algorithm using a MEE‐like concept is also presented. Examples, tests, evaluation experiments and comparison with similar machines using classic approaches, complement the descriptions.
Classification error of the thresholded independence rule
DEFF Research Database (Denmark)
Bak, Britta Anker; Fenger-Grøn, Morten; Jensen, Jens Ledet
We consider classification in the situation of two groups with normally distributed data in the ‘large p small n’ framework. To counterbalance the high number of variables we consider the thresholded independence rule. An upper bound on the classification error is established which is taylored...
Directory of Open Access Journals (Sweden)
Zhigao Zeng
2016-01-01
Full Text Available This paper proposes a novel algorithm to solve the challenging problem of classifying error-diffused halftone images. We firstly design the class feature matrices, after extracting the image patches according to their statistics characteristics, to classify the error-diffused halftone images. Then, the spectral regression kernel discriminant analysis is used for feature dimension reduction. The error-diffused halftone images are finally classified using an idea similar to the nearest centroids classifier. As demonstrated by the experimental results, our method is fast and can achieve a high classification accuracy rate with an added benefit of robustness in tackling noise.
Error Detection and Error Classification: Failure Awareness in Data Transfer Scheduling
Energy Technology Data Exchange (ETDEWEB)
Louisiana State University; Balman, Mehmet; Kosar, Tevfik
2010-10-27
Data transfer in distributed environment is prone to frequent failures resulting from back-end system level problems, like connectivity failure which is technically untraceable by users. Error messages are not logged efficiently, and sometimes are not relevant/useful from users point-of-view. Our study explores the possibility of an efficient error detection and reporting system for such environments. Prior knowledge about the environment and awareness of the actual reason behind a failure would enable higher level planners to make better and accurate decisions. It is necessary to have well defined error detection and error reporting methods to increase the usability and serviceability of existing data transfer protocols and data management systems. We investigate the applicability of early error detection and error classification techniques and propose an error reporting framework and a failure-aware data transfer life cycle to improve arrangement of data transfer operations and to enhance decision making of data transfer schedulers.
Automated Classification of Phonological Errors in Aphasic Language
Ahuja, Sanjeev B.; Reggia, James A.; Berndt, Rita S.
1984-01-01
Using heuristically-guided state space search, a prototype program has been developed to simulate and classify phonemic errors occurring in the speech of neurologically-impaired patients. Simulations are based on an interchangeable rule/operator set of elementary errors which represent a theory of phonemic processing faults. This work introduces and evaluates a novel approach to error simulation and classification, it provides a prototype simulation tool for neurolinguistic research, and it forms the initial phase of a larger research effort involving computer modelling of neurolinguistic processes.
Medication errors: definitions and classification
Aronson, Jeffrey K
2009-01-01
To understand medication errors and to identify preventive strategies, we need to classify them and define the terms that describe them. The four main approaches to defining technical terms consider etymology, usage, previous definitions, and the Ramsey–Lewis method (based on an understanding of theory and practice). A medication error is ‘a failure in the treatment process that leads to, or has the potential to lead to, harm to the patient’. Prescribing faults, a subset of medication errors, should be distinguished from prescription errors. A prescribing fault is ‘a failure in the prescribing [decision-making] process that leads to, or has the potential to lead to, harm to the patient’. The converse of this, ‘balanced prescribing’ is ‘the use of a medicine that is appropriate to the patient's condition and, within the limits created by the uncertainty that attends therapeutic decisions, in a dosage regimen that optimizes the balance of benefit to harm’. This excludes all forms of prescribing faults, such as irrational, inappropriate, and ineffective prescribing, underprescribing and overprescribing. A prescription error is ‘a failure in the prescription writing process that results in a wrong instruction about one or more of the normal features of a prescription’. The ‘normal features’ include the identity of the recipient, the identity of the drug, the formulation, dose, route, timing, frequency, and duration of administration. Medication errors can be classified, invoking psychological theory, as knowledge-based mistakes, rule-based mistakes, action-based slips, and memory-based lapses. This classification informs preventive strategies. PMID:19594526
Directory of Open Access Journals (Sweden)
Kamil Zidek
2016-10-01
Full Text Available The article deals with the design of embedded vision equipment of industrial robots for inline diagnosis of product error during manipulation process. The vision equipment can be attached to the end effector of robots or manipulators, and it provides an image snapshot of part surface before grasp, searches for error during manipulation, and separates products with error from the next operation of manufacturing. The new approach is a methodology based on machine teaching for the automated identification, localization, and diagnosis of systematic errors in products of high-volume production. To achieve this, we used two main data mining algorithms: clustering for accumulation of similar errors and classification methods for the prediction of any new error to proposed class. The presented methodology consists of three separate processing levels: image acquisition for fail parameterization, data clustering for categorizing errors to separate classes, and new pattern prediction with a proposed class model. We choose main representatives of clustering algorithms, for example, K-mean from quantization of vectors, fast library for approximate nearest neighbor from hierarchical clustering, and density-based spatial clustering of applications with noise from algorithm based on the density of the data. For machine learning, we selected six major algorithms of classification: support vector machines, normal Bayesian classifier, K-nearest neighbor, gradient boosted trees, random trees, and neural networks. The selected algorithms were compared for speed and reliability and tested on two platforms: desktop-based computer system and embedded system based on System on Chip (SoC with vision equipment.
Pitch Based Sound Classification
DEFF Research Database (Denmark)
Nielsen, Andreas Brinch; Hansen, Lars Kai; Kjems, U
2006-01-01
A sound classification model is presented that can classify signals into music, noise and speech. The model extracts the pitch of the signal using the harmonic product spectrum. Based on the pitch estimate and a pitch error measure, features are created and used in a probabilistic model with soft......-max output function. Both linear and quadratic inputs are used. The model is trained on 2 hours of sound and tested on publicly available data. A test classification error below 0.05 with 1 s classification windows is achieved. Further more it is shown that linear input performs as well as a quadratic......, and that even though classification gets marginally better, not much is achieved by increasing the window size beyond 1 s....
Cohen, Aaron M
2008-01-01
We participated in the i2b2 smoking status classification challenge task. The purpose of this task was to evaluate the ability of systems to automatically identify patient smoking status from discharge summaries. Our submission included several techniques that we compared and studied, including hot-spot identification, zero-vector filtering, inverse class frequency weighting, error-correcting output codes, and post-processing rules. We evaluated our approaches using the same methods as the i2b2 task organizers, using micro- and macro-averaged F1 as the primary performance metric. Our best performing system achieved a micro-F1 of 0.9000 on the test collection, equivalent to the best performing system submitted to the i2b2 challenge. Hot-spot identification, zero-vector filtering, classifier weighting, and error correcting output coding contributed additively to increased performance, with hot-spot identification having by far the largest positive effect. High performance on automatic identification of patient smoking status from discharge summaries is achievable with the efficient and straightforward machine learning techniques studied here.
Boeschoten, Laura; Oberski, Daniel; De Waal, Ton
2017-01-01
Both registers and surveys can contain classification errors. These errors can be estimated by making use of a composite data set. We propose a new method based on latent class modelling to estimate the number of classification errors across several sources while taking into account impossible
Convolutional Neural Network Achieves Human-level Accuracy in Music Genre Classification
Dong, Mingwen
2018-01-01
Music genre classification is one example of content-based analysis of music signals. Traditionally, human-engineered features were used to automatize this task and 61% accuracy has been achieved in the 10-genre classification. However, it's still below the 70% accuracy that humans could achieve in the same task. Here, we propose a new method that combines knowledge of human perception study in music genre classification and the neurophysiology of the auditory system. The method works by trai...
Classification based upon gene expression data: bias and precision of error rates.
Wood, Ian A; Visscher, Peter M; Mengersen, Kerrie L
2007-06-01
Gene expression data offer a large number of potentially useful predictors for the classification of tissue samples into classes, such as diseased and non-diseased. The predictive error rate of classifiers can be estimated using methods such as cross-validation. We have investigated issues of interpretation and potential bias in the reporting of error rate estimates. The issues considered here are optimization and selection biases, sampling effects, measures of misclassification rate, baseline error rates, two-level external cross-validation and a novel proposal for detection of bias using the permutation mean. Reporting an optimal estimated error rate incurs an optimization bias. Downward bias of 3-5% was found in an existing study of classification based on gene expression data and may be endemic in similar studies. Using a simulated non-informative dataset and two example datasets from existing studies, we show how bias can be detected through the use of label permutations and avoided using two-level external cross-validation. Some studies avoid optimization bias by using single-level cross-validation and a test set, but error rates can be more accurately estimated via two-level cross-validation. In addition to estimating the simple overall error rate, we recommend reporting class error rates plus where possible the conditional risk incorporating prior class probabilities and a misclassification cost matrix. We also describe baseline error rates derived from three trivial classifiers which ignore the predictors. R code which implements two-level external cross-validation with the PAMR package, experiment code, dataset details and additional figures are freely available for non-commercial use from http://www.maths.qut.edu.au/profiles/wood/permr.jsp
Hooper, Brionny J; O'Hare, David P A
2013-08-01
Human error classification systems theoretically allow researchers to analyze postaccident data in an objective and consistent manner. The Human Factors Analysis and Classification System (HFACS) framework is one such practical analysis tool that has been widely used to classify human error in aviation. The Cognitive Error Taxonomy (CET) is another. It has been postulated that the focus on interrelationships within HFACS can facilitate the identification of the underlying causes of pilot error. The CET provides increased granularity at the level of unsafe acts. The aim was to analyze the influence of factors at higher organizational levels on the unsafe acts of front-line operators and to compare the errors of fixed-wing and rotary-wing operations. This study analyzed 288 aircraft incidents involving human error from an Australasian military organization occurring between 2001 and 2008. Action errors accounted for almost twice (44%) the proportion of rotary wing compared to fixed wing (23%) incidents. Both classificatory systems showed significant relationships between precursor factors such as the physical environment, mental and physiological states, crew resource management, training and personal readiness, and skill-based, but not decision-based, acts. The CET analysis showed different predisposing factors for different aspects of skill-based behaviors. Skill-based errors in military operations are more prevalent in rotary wing incidents and are related to higher level supervisory processes in the organization. The Cognitive Error Taxonomy provides increased granularity to HFACS analyses of unsafe acts.
Directory of Open Access Journals (Sweden)
Xiao-Ping Zhu
2014-09-01
Conclusion: Application of this classification system will help nursing administrators to accurately detect system- and process-related defects leading to medication errors, and enable the factors to be targeted to improve the level of patient safety management.
Early math and reading achievement are associated with the error positivity
Directory of Open Access Journals (Sweden)
Matthew H. Kim
2016-12-01
Full Text Available Executive functioning (EF and motivation are associated with academic achievement and error-related ERPs. The present study explores whether early academic skills predict variability in the error-related negativity (ERN and error positivity (Pe. Data from 113 three- to seven-year-old children in a Go/No-Go task revealed that stronger early reading and math skills predicted a larger Pe. Closer examination revealed that this relation was quadratic and significant for children performing at or near grade level, but not significant for above-average achievers. Early academics did not predict the ERN. These findings suggest that the Pe – which reflects individual differences in motivational processes as well as attention – may be associated with early academic achievement.
Evaluating Method Engineer Performance: an error classification and preliminary empirical study
Directory of Open Access Journals (Sweden)
Steven Kelly
1998-11-01
Full Text Available We describe an approach to empirically test the use of metaCASE environments to model methods. Both diagrams and matrices have been proposed as a means for presenting the methods. These different paradigms may have their own effects on how easily and well users can model methods. We extend Batra's classification of errors in data modelling to cover metamodelling, and use it to measure the performance of a group of metamodellers using either diagrams or matrices. The tentative results from this pilot study confirm the usefulness of the classification, and show some interesting differences between the paradigms.
Fully Convolutional Networks for Ground Classification from LIDAR Point Clouds
Rizaldy, A.; Persello, C.; Gevaert, C. M.; Oude Elberink, S. J.
2018-05-01
Deep Learning has been massively used for image classification in recent years. The use of deep learning for ground classification from LIDAR point clouds has also been recently studied. However, point clouds need to be converted into an image in order to use Convolutional Neural Networks (CNNs). In state-of-the-art techniques, this conversion is slow because each point is converted into a separate image. This approach leads to highly redundant computation during conversion and classification. The goal of this study is to design a more efficient data conversion and ground classification. This goal is achieved by first converting the whole point cloud into a single image. The classification is then performed by a Fully Convolutional Network (FCN), a modified version of CNN designed for pixel-wise image classification. The proposed method is significantly faster than state-of-the-art techniques. On the ISPRS Filter Test dataset, it is 78 times faster for conversion and 16 times faster for classification. Our experimental analysis on the same dataset shows that the proposed method results in 5.22 % of total error, 4.10 % of type I error, and 15.07 % of type II error. Compared to the previous CNN-based technique and LAStools software, the proposed method reduces the total error and type I error (while type II error is slightly higher). The method was also tested on a very high point density LIDAR point clouds resulting in 4.02 % of total error, 2.15 % of type I error and 6.14 % of type II error.
Achieving the Heisenberg limit in quantum metrology using quantum error correction.
Zhou, Sisi; Zhang, Mengzhen; Preskill, John; Jiang, Liang
2018-01-08
Quantum metrology has many important applications in science and technology, ranging from frequency spectroscopy to gravitational wave detection. Quantum mechanics imposes a fundamental limit on measurement precision, called the Heisenberg limit, which can be achieved for noiseless quantum systems, but is not achievable in general for systems subject to noise. Here we study how measurement precision can be enhanced through quantum error correction, a general method for protecting a quantum system from the damaging effects of noise. We find a necessary and sufficient condition for achieving the Heisenberg limit using quantum probes subject to Markovian noise, assuming that noiseless ancilla systems are available, and that fast, accurate quantum processing can be performed. When the sufficient condition is satisfied, a quantum error-correcting code can be constructed that suppresses the noise without obscuring the signal; the optimal code, achieving the best possible precision, can be found by solving a semidefinite program.
A neural network for noise correlation classification
Paitz, Patrick; Gokhberg, Alexey; Fichtner, Andreas
2018-02-01
We present an artificial neural network (ANN) for the classification of ambient seismic noise correlations into two categories, suitable and unsuitable for noise tomography. By using only a small manually classified data subset for network training, the ANN allows us to classify large data volumes with low human effort and to encode the valuable subjective experience of data analysts that cannot be captured by a deterministic algorithm. Based on a new feature extraction procedure that exploits the wavelet-like nature of seismic time-series, we efficiently reduce the dimensionality of noise correlation data, still keeping relevant features needed for automated classification. Using global- and regional-scale data sets, we show that classification errors of 20 per cent or less can be achieved when the network training is performed with as little as 3.5 per cent and 16 per cent of the data sets, respectively. Furthermore, the ANN trained on the regional data can be applied to the global data, and vice versa, without a significant increase of the classification error. An experiment where four students manually classified the data, revealed that the classification error they would assign to each other is substantially larger than the classification error of the ANN (>35 per cent). This indicates that reproducibility would be hampered more by human subjectivity than by imperfections of the ANN.
FULLY CONVOLUTIONAL NETWORKS FOR GROUND CLASSIFICATION FROM LIDAR POINT CLOUDS
Directory of Open Access Journals (Sweden)
A. Rizaldy
2018-05-01
Full Text Available Deep Learning has been massively used for image classification in recent years. The use of deep learning for ground classification from LIDAR point clouds has also been recently studied. However, point clouds need to be converted into an image in order to use Convolutional Neural Networks (CNNs. In state-of-the-art techniques, this conversion is slow because each point is converted into a separate image. This approach leads to highly redundant computation during conversion and classification. The goal of this study is to design a more efficient data conversion and ground classification. This goal is achieved by first converting the whole point cloud into a single image. The classification is then performed by a Fully Convolutional Network (FCN, a modified version of CNN designed for pixel-wise image classification. The proposed method is significantly faster than state-of-the-art techniques. On the ISPRS Filter Test dataset, it is 78 times faster for conversion and 16 times faster for classification. Our experimental analysis on the same dataset shows that the proposed method results in 5.22 % of total error, 4.10 % of type I error, and 15.07 % of type II error. Compared to the previous CNN-based technique and LAStools software, the proposed method reduces the total error and type I error (while type II error is slightly higher. The method was also tested on a very high point density LIDAR point clouds resulting in 4.02 % of total error, 2.15 % of type I error and 6.14 % of type II error.
The Influence of Guided Error-Based Learning on Motor Skills Self-Efficacy and Achievement.
Chien, Kuei-Pin; Chen, Sufen
2018-01-01
The authors investigated the role of errors in motor skills teaching, specifically the influence of errors on skills self-efficacy and achievement. The participants were 75 undergraduate students enrolled in pétanque courses. The experimental group (guided error-based learning, n = 37) received a 6-week period of instruction based on the students' errors, whereas the control group (correct motion instruction, n = 38) received a 6-week period of instruction emphasizing correct motor skills. The experimental group had significantly higher scores in motor skills self-efficacy and outcomes than did the control group. Novices' errors reflect their schema in motor skills learning, which provides a basis for instructors to implement student-centered instruction and to facilitate the learning process. Guided error-based learning can effectively enhance beginners' skills self-efficacy and achievement in precision sports such as pétanque.
Directory of Open Access Journals (Sweden)
Boeschoten Laura
2017-12-01
Full Text Available Both registers and surveys can contain classification errors. These errors can be estimated by making use of a composite data set. We propose a new method based on latent class modelling to estimate the number of classification errors across several sources while taking into account impossible combinations with scores on other variables. Furthermore, the latent class model, by multiply imputing a new variable, enhances the quality of statistics based on the composite data set. The performance of this method is investigated by a simulation study, which shows that whether or not the method can be applied depends on the entropy R2 of the latent class model and the type of analysis a researcher is planning to do. Finally, the method is applied to public data from Statistics Netherlands.
Selectively Fortifying Reconfigurable Computing Device to Achieve Higher Error Resilience
Directory of Open Access Journals (Sweden)
Mingjie Lin
2012-01-01
Full Text Available With the advent of 10 nm CMOS devices and “exotic” nanodevices, the location and occurrence time of hardware defects and design faults become increasingly unpredictable, therefore posing severe challenges to existing techniques for error-resilient computing because most of them statically assign hardware redundancy and do not account for the error tolerance inherently existing in many mission-critical applications. This work proposes a novel approach to selectively fortifying a target reconfigurable computing device in order to achieve hardware-efficient error resilience for a specific target application. We intend to demonstrate that such error resilience can be significantly improved with effective hardware support. The major contributions of this work include (1 the development of a complete methodology to perform sensitivity and criticality analysis of hardware redundancy, (2 a novel problem formulation and an efficient heuristic methodology to selectively allocate hardware redundancy among a target design’s key components in order to maximize its overall error resilience, and (3 an academic prototype of SFC computing device that illustrates a 4 times improvement of error resilience for a H.264 encoder implemented with an FPGA device.
Software platform for managing the classification of error- related potentials of observers
Asvestas, P.; Ventouras, E.-C.; Kostopoulos, S.; Sidiropoulos, K.; Korfiatis, V.; Korda, A.; Uzunolglu, A.; Karanasiou, I.; Kalatzis, I.; Matsopoulos, G.
2015-09-01
Human learning is partly based on observation. Electroencephalographic recordings of subjects who perform acts (actors) or observe actors (observers), contain a negative waveform in the Evoked Potentials (EPs) of the actors that commit errors and of observers who observe the error-committing actors. This waveform is called the Error-Related Negativity (ERN). Its detection has applications in the context of Brain-Computer Interfaces. The present work describes a software system developed for managing EPs of observers, with the aim of classifying them into observations of either correct or incorrect actions. It consists of an integrated platform for the storage, management, processing and classification of EPs recorded during error-observation experiments. The system was developed using C# and the following development tools and frameworks: MySQL, .NET Framework, Entity Framework and Emgu CV, for interfacing with the machine learning library of OpenCV. Up to six features can be computed per EP recording per electrode. The user can select among various feature selection algorithms and then proceed to train one of three types of classifiers: Artificial Neural Networks, Support Vector Machines, k-nearest neighbour. Next the classifier can be used for classifying any EP curve that has been inputted to the database.
Maximum mutual information regularized classification
Wang, Jim Jing-Yan
2014-09-07
In this paper, a novel pattern classification approach is proposed by regularizing the classifier learning to maximize mutual information between the classification response and the true class label. We argue that, with the learned classifier, the uncertainty of the true class label of a data sample should be reduced by knowing its classification response as much as possible. The reduced uncertainty is measured by the mutual information between the classification response and the true class label. To this end, when learning a linear classifier, we propose to maximize the mutual information between classification responses and true class labels of training samples, besides minimizing the classification error and reducing the classifier complexity. An objective function is constructed by modeling mutual information with entropy estimation, and it is optimized by a gradient descend method in an iterative algorithm. Experiments on two real world pattern classification problems show the significant improvements achieved by maximum mutual information regularization.
Maximum mutual information regularized classification
Wang, Jim Jing-Yan; Wang, Yi; Zhao, Shiguang; Gao, Xin
2014-01-01
In this paper, a novel pattern classification approach is proposed by regularizing the classifier learning to maximize mutual information between the classification response and the true class label. We argue that, with the learned classifier, the uncertainty of the true class label of a data sample should be reduced by knowing its classification response as much as possible. The reduced uncertainty is measured by the mutual information between the classification response and the true class label. To this end, when learning a linear classifier, we propose to maximize the mutual information between classification responses and true class labels of training samples, besides minimizing the classification error and reducing the classifier complexity. An objective function is constructed by modeling mutual information with entropy estimation, and it is optimized by a gradient descend method in an iterative algorithm. Experiments on two real world pattern classification problems show the significant improvements achieved by maximum mutual information regularization.
Development of a methodology for classifying software errors
Gerhart, S. L.
1976-01-01
A mathematical formalization of the intuition behind classification of software errors is devised and then extended to a classification discipline: Every classification scheme should have an easily discernible mathematical structure and certain properties of the scheme should be decidable (although whether or not these properties hold is relative to the intended use of the scheme). Classification of errors then becomes an iterative process of generalization from actual errors to terms defining the errors together with adjustment of definitions according to the classification discipline. Alternatively, whenever possible, small scale models may be built to give more substance to the definitions. The classification discipline and the difficulties of definition are illustrated by examples of classification schemes from the literature and a new study of observed errors in published papers of programming methodologies.
Maslanik, J. A.; Key, J.
1992-01-01
An expert system framework has been developed to classify sea ice types using satellite passive microwave data, an operational classification algorithm, spatial and temporal information, ice types estimated from a dynamic-thermodynamic model, output from a neural network that detects the onset of melt, and knowledge about season and region. The rule base imposes boundary conditions upon the ice classification, modifies parameters in the ice algorithm, determines a `confidence' measure for the classified data, and under certain conditions, replaces the algorithm output with model output. Results demonstrate the potential power of such a system for minimizing overall error in the classification and for providing non-expert data users with a means of assessing the usefulness of the classification results for their applications.
One-Class Classification-Based Real-Time Activity Error Detection in Smart Homes.
Das, Barnan; Cook, Diane J; Krishnan, Narayanan C; Schmitter-Edgecombe, Maureen
2016-08-01
Caring for individuals with dementia is frequently associated with extreme physical and emotional stress, which often leads to depression. Smart home technology and advances in machine learning techniques can provide innovative solutions to reduce caregiver burden. One key service that caregivers provide is prompting individuals with memory limitations to initiate and complete daily activities. We hypothesize that sensor technologies combined with machine learning techniques can automate the process of providing reminder-based interventions. The first step towards automated interventions is to detect when an individual faces difficulty with activities. We propose machine learning approaches based on one-class classification that learn normal activity patterns. When we apply these classifiers to activity patterns that were not seen before, the classifiers are able to detect activity errors, which represent potential prompt situations. We validate our approaches on smart home sensor data obtained from older adult participants, some of whom faced difficulties performing routine activities and thus committed errors.
Schumacher, Robin F; Malone, Amelia S
2017-09-01
The goal of the present study was to describe fraction-calculation errors among 4 th -grade students and determine whether error patterns differed as a function of problem type (addition vs. subtraction; like vs. unlike denominators), orientation (horizontal vs. vertical), or mathematics-achievement status (low- vs. average- vs. high-achieving). We specifically addressed whether mathematics-achievement status was related to students' tendency to operate with whole number bias. We extended this focus by comparing low-performing students' errors in two instructional settings that focused on two different types of fraction understandings: core instruction that focused on part-whole understanding vs. small-group tutoring that focused on magnitude understanding. Results showed students across the sample were more likely to operate with whole number bias on problems with unlike denominators. Students with low or average achievement (who only participated in core instruction) were more likely to operate with whole number bias than students with low achievement who participated in small-group tutoring. We suggest instruction should emphasize magnitude understanding to sufficiently increase fraction understanding for all students in the upper elementary grades.
Directory of Open Access Journals (Sweden)
Hendrik eMandelkow
2016-03-01
Full Text Available Naturalistic stimuli like movies evoke complex perceptual processes, which are of great interest in the study of human cognition by functional MRI (fMRI. However, conventional fMRI analysis based on statistical parametric mapping (SPM and the general linear model (GLM is hampered by a lack of accurate parametric models of the BOLD response to complex stimuli. In this situation, statistical machine-learning methods, a.k.a. multivariate pattern analysis (MVPA, have received growing attention for their ability to generate stimulus response models in a data-driven fashion. However, machine-learning methods typically require large amounts of training data as well as computational resources. In the past this has largely limited their application to fMRI experiments involving small sets of stimulus categories and small regions of interest in the brain. By contrast, the present study compares several classification algorithms known as Nearest Neighbour (NN, Gaussian Naïve Bayes (GNB, and (regularised Linear Discriminant Analysis (LDA in terms of their classification accuracy in discriminating the global fMRI response patterns evoked by a large number of naturalistic visual stimuli presented as a movie.Results show that LDA regularised by principal component analysis (PCA achieved high classification accuracies, above 90% on average for single fMRI volumes acquired 2s apart during a 300s movie (chance level 0.7% = 2s/300s. The largest source of classification errors were autocorrelations in the BOLD signal compounded by the similarity of consecutive stimuli. All classifiers performed best when given input features from a large region of interest comprising around 25% of the voxels that responded significantly to the visual stimulus. Consistent with this, the most informative principal components represented widespread distributions of co-activated brain regions that were similar between subjects and may represent functional networks. In light of these
International Nuclear Information System (INIS)
Takagawa, Kenichi
2004-01-01
At nuclear power plants, plant operators examine other human error cases, including those that occurred at other plants, so that they can learn from such experiences and avoid making similar errors again. Although there is little data available on errors made at domestic plants, nuclear operators in foreign countries are reporting even minor irregularities and signs of faults, and a large amount of data on human errors at overseas plants could be collected and examined. However, these overseas data have not been used effectively because most of them are poorly organized or not properly classified and are often hard to understand. Accordingly, we carried out a study on the cases of human errors at overseas power plants in order to help plant personnel clearly understand overseas experiences and avoid repeating similar errors, The study produced the following results, which were put to use at nuclear power plants and other facilities. (1) ''One-Point-Advice'' refers to a practice where a leader gives pieces of advice to his team of operators in order to prevent human errors before starting work. Based on this practice and those used in the aviation industry, we have developed a new method of classifying human errors that consists of four basic actions and three applied actions. (2) We used this new classification method to classify human errors made by operators at overseas nuclear power plants. The results show that the most frequent errors caused not by operators themselves but due to insufficient team monitoring, for which superiors and/or their colleagues were responsible. We therefore analyzed and classified possible factors contributing to insufficient team monitoring, and demonstrated that the frequent errors have also occurred at domestic power plants. (3) Using the new classification formula, we prepared a human error case sheets that is easy for plant personnel to understand. The sheets are designed to make data more understandable and easier to remember
Cohen, Aaron M.
2008-01-01
We participated in the i2b2 smoking status classification challenge task. The purpose of this task was to evaluate the ability of systems to automatically identify patient smoking status from discharge summaries. Our submission included several techniques that we compared and studied, including hot-spot identification, zero-vector filtering, inverse class frequency weighting, error-correcting output codes, and post-processing rules. We evaluated our approaches using the same methods as the i2...
Error studies of Halbach Magnets
Energy Technology Data Exchange (ETDEWEB)
Brooks, S. [Brookhaven National Lab. (BNL), Upton, NY (United States)
2017-03-02
These error studies were done on the Halbach magnets for the CBETA “First Girder” as described in note [CBETA001]. The CBETA magnets have since changed slightly to the lattice in [CBETA009]. However, this is not a large enough change to significantly affect the results here. The QF and BD arc FFAG magnets are considered. For each assumed set of error distributions and each ideal magnet, 100 random magnets with errors are generated. These are then run through an automated version of the iron wire multipole cancellation algorithm. The maximum wire diameter allowed is 0.063” as in the proof-of-principle magnets. Initially, 32 wires (2 per Halbach wedge) are tried, then if this does not achieve 1e-4 level accuracy in the simulation, 48 and then 64 wires. By “1e-4 accuracy”, it is meant the FOM defined by √(Σn≥sextupole an 2+bn 2) is less than 1 unit, where the multipoles are taken at the maximum nominal beam radius, R=23mm for these magnets. The algorithm initially uses 20 convergence interations. If 64 wires does not achieve 1e-4 accuracy, this is increased to 50 iterations to check for slow converging cases. There are also classifications for magnets that do not achieve 1e-4 but do achieve 1e-3 (FOM ≤ 10 units). This is technically within the spec discussed in the Jan 30, 2017 review; however, there will be errors in practical shimming not dealt with in the simulation, so it is preferable to do much better than the spec in the simulation.
Waade, Gunvor Gipling; Highnam, Ralph; Hauge, Ingrid H R; McEntee, Mark F; Hofvind, Solveig; Denton, Erika; Kelly, Judith; Sarwar, Jasmine J; Hogg, Peter
2016-06-01
Mammographic density has been demonstrated to predict breast cancer risk. It has been proposed that it could be used for stratifying screening pathways and recommending additional imaging. Volumetric density tools use the recorded compressed breast thickness (CBT) of the breast measured at the x-ray unit in their calculation; however, the accuracy of the recorded thickness can vary. The aim of this study was to investigate whether inaccuracies in recorded CBT impact upon volumetric density classification and to examine whether the current quality control (QC) standard is sufficient for assessing mammographic density. Raw data from 52 digital screening mammograms were included in the study. For each image, the clinically recorded CBT was artificially increased and decreased in increments of 1 mm to simulate measurement error, until ±15% from the recorded CBT was reached. New images were created for each 1 mm step in thickness resulting in a total of 974 images which then had volpara density grade (VDG) and volumetric density percentage assigned. A change in VDG was observed in 38.5% (n = 20) of mammograms when applying ±15% error to the recorded CBT and 11.5% (n = 6) was within the QC standard prescribed error of ±5 mm. The current QC standard of ±5 mm error in recorded CBT creates the potential for error in mammographic density measurement. This may lead to inaccurate classification of mammographic density. The current QC standard for assessing mammographic density should be reconsidered.
Directory of Open Access Journals (Sweden)
Astri J Lundervold
Full Text Available Inattention in childhood is associated with academic problems later in life. The contribution of specific aspects of inattentive behaviour is, however, less known. We investigated feature importance of primary school teachers' reports on nine aspects of inattentive behaviour, gender and age in predicting future academic achievement. Primary school teachers of n = 2491 children (7-9 years rated nine items reflecting different aspects of inattentive behaviour in 2002. A mean academic achievement score from the previous semester in high school (2012 was available for each youth from an official school register. All scores were at a categorical level. Feature importances were assessed by using multinominal logistic regression, classification and regression trees analysis, and a random forest algorithm. Finally, a comprehensive pattern classification procedure using k-fold cross-validation was implemented. Overall, inattention was rated as more severe in boys, who also obtained lower academic achievement scores in high school than girls. Problems related to sustained attention and distractibility were together with age and gender defined as the most important features to predict future achievement scores. Using these four features as input to a collection of classifiers employing k-fold cross-validation for prediction of academic achievement level, we obtained classification accuracy, precision and recall that were clearly better than chance levels. Primary school teachers' reports of problems related to sustained attention and distractibility were identified as the two most important features of inattentive behaviour predicting academic achievement in high school. Identification and follow-up procedures of primary school children showing these characteristics should be prioritised to prevent future academic failure.
Lundervold, Astri J; Bøe, Tormod; Lundervold, Arvid
2017-01-01
Inattention in childhood is associated with academic problems later in life. The contribution of specific aspects of inattentive behaviour is, however, less known. We investigated feature importance of primary school teachers' reports on nine aspects of inattentive behaviour, gender and age in predicting future academic achievement. Primary school teachers of n = 2491 children (7-9 years) rated nine items reflecting different aspects of inattentive behaviour in 2002. A mean academic achievement score from the previous semester in high school (2012) was available for each youth from an official school register. All scores were at a categorical level. Feature importances were assessed by using multinominal logistic regression, classification and regression trees analysis, and a random forest algorithm. Finally, a comprehensive pattern classification procedure using k-fold cross-validation was implemented. Overall, inattention was rated as more severe in boys, who also obtained lower academic achievement scores in high school than girls. Problems related to sustained attention and distractibility were together with age and gender defined as the most important features to predict future achievement scores. Using these four features as input to a collection of classifiers employing k-fold cross-validation for prediction of academic achievement level, we obtained classification accuracy, precision and recall that were clearly better than chance levels. Primary school teachers' reports of problems related to sustained attention and distractibility were identified as the two most important features of inattentive behaviour predicting academic achievement in high school. Identification and follow-up procedures of primary school children showing these characteristics should be prioritised to prevent future academic failure.
Sakuma, Jun; Wright, Rebecca N.
Privacy-preserving classification is the task of learning or training a classifier on the union of privately distributed datasets without sharing the datasets. The emphasis of existing studies in privacy-preserving classification has primarily been put on the design of privacy-preserving versions of particular data mining algorithms, However, in classification problems, preprocessing and postprocessing— such as model selection or attribute selection—play a prominent role in achieving higher classification accuracy. In this paper, we show generalization error of classifiers in privacy-preserving classification can be securely evaluated without sharing prediction results. Our main technical contribution is a new generalized Hamming distance protocol that is universally applicable to preprocessing and postprocessing of various privacy-preserving classification problems, such as model selection in support vector machine and attribute selection in naive Bayes classification.
3D multi-view convolutional neural networks for lung nodule classification
Kang, Guixia; Hou, Beibei; Zhang, Ningbo
2017-01-01
The 3D convolutional neural network (CNN) is able to make full use of the spatial 3D context information of lung nodules, and the multi-view strategy has been shown to be useful for improving the performance of 2D CNN in classifying lung nodules. In this paper, we explore the classification of lung nodules using the 3D multi-view convolutional neural networks (MV-CNN) with both chain architecture and directed acyclic graph architecture, including 3D Inception and 3D Inception-ResNet. All networks employ the multi-view-one-network strategy. We conduct a binary classification (benign and malignant) and a ternary classification (benign, primary malignant and metastatic malignant) on Computed Tomography (CT) images from Lung Image Database Consortium and Image Database Resource Initiative database (LIDC-IDRI). All results are obtained via 10-fold cross validation. As regards the MV-CNN with chain architecture, results show that the performance of 3D MV-CNN surpasses that of 2D MV-CNN by a significant margin. Finally, a 3D Inception network achieved an error rate of 4.59% for the binary classification and 7.70% for the ternary classification, both of which represent superior results for the corresponding task. We compare the multi-view-one-network strategy with the one-view-one-network strategy. The results reveal that the multi-view-one-network strategy can achieve a lower error rate than the one-view-one-network strategy. PMID:29145492
International Nuclear Information System (INIS)
Miyazaki, Takamasa
2007-01-01
The adverse events which occurred in nuclear power plants are analyzed to prevent similar events, and in the analysis of each event, the cause of the event is classified by a cause classification method. This paper shows a new cause classification method which is improved in several points as follows: (1) the whole causes are systematically classified into three major categories such as machine system, operation system and plant outside causes, (2) the causes of the operation system are classified into several management errors normally performed in a nuclear power plant, (3) the content of ageing is defined in detail for their further analysis, (4) human errors are divided and defined by the error stage, (5) human errors can be related to background factors, and so on. This new method is applied to the adverse events which occurred in domestic and overseas nuclear power plants in 2005. From these results, it is clarified that operation system errors account for about 60% of the whole causes, of which approximately 60% are maintenance errors, about 40% are worker's human errors, and that the prevention of maintenance errors, especially worker's human errors is crucial. (author)
Signal classification for acoustic neutrino detection
International Nuclear Information System (INIS)
Neff, M.; Anton, G.; Enzenhöfer, A.; Graf, K.; Hößl, J.; Katz, U.; Lahmann, R.; Richardt, C.
2012-01-01
This article focuses on signal classification for deep-sea acoustic neutrino detection. In the deep sea, the background of transient signals is very diverse. Approaches like matched filtering are not sufficient to distinguish between neutrino-like signals and other transient signals with similar signature, which are forming the acoustic background for neutrino detection in the deep-sea environment. A classification system based on machine learning algorithms is analysed with the goal to find a robust and effective way to perform this task. For a well-trained model, a testing error on the level of 1% is achieved for strong classifiers like Random Forest and Boosting Trees using the extracted features of the signal as input and utilising dense clusters of sensors instead of single sensors.
Directory of Open Access Journals (Sweden)
Moh. Aries Syufagi
2012-07-01
Full Text Available To assess the cognitive level of player ability is difficult; many instruments are potentially biased, unreliable, and invalid test. Whereas, in serious game is important to know the cognitive level. If the cognitive level can be measured well, the mastery learning can be achieved. Mastery learning is the core of the learning process in serious game. To classify the cognitive level of players, researchers propose a Cognitive Skill Game (CSG. CSG improves this cognitive concept to monitor how players interact with the game. This game employs Learning Vector Quantization (LVQ for optimizing the cognitive skill input classification of the player. Training data in LVQ use data observation from the teacher. Populations of cognitive skill classification in this research are pupils when playing the game. Mostly players cognitive skill game have cognitive skill category are Trial and Error. Some of them have Expert category, and a few included in the group carefully. Thus, the general level of skill of the player is still low. Untuk menilai tingkat kognitif dari kemampuan pemain sangatlah sulit; banyak instrumen yang berpotensi bias, tidak dapat diandalkan, dan merupakan tes yang tidak valid. Padahal, dalam serious game penting untuk mengetahui tingkat kognitif. Jika tingkat kognitif dapat diukur dengan baik, penguasaan pembelajaran dapat dicapai. Penguasaan belajar adalah inti dari proses belajar dalam serious game. Untuk mengklasifikasikan tingkat kognitif pemain, kami mengusulkan Cognitive Skill Game (CSG. CSG meningkatkan konsep kognitif untuk memantau bagaimana pemain berinteraksi dengan permainan. Permainan ini menggunakan Learning Vector Quantization (LVQ untuk mengoptimalkan input klasifikasi keterampilan kognitif pemain. Data trining dalam observasi LVQ menggunakan data dari guru. Populasi klasifikasi keterampilan kognitif dalam penelitian ini adalah siswa saat memainkan permainan. Sebagian besar pemain CSG berkategori keterampilan kognitif
High-Performance Neural Networks for Visual Object Classification
Cireşan, Dan C.; Meier, Ueli; Masci, Jonathan; Gambardella, Luca M.; Schmidhuber, Jürgen
2011-01-01
We present a fast, fully parameterizable GPU implementation of Convolutional Neural Network variants. Our feature extractors are neither carefully designed nor pre-wired, but rather learned in a supervised way. Our deep hierarchical architectures achieve the best published results on benchmarks for object classification (NORB, CIFAR10) and handwritten digit recognition (MNIST), with error rates of 2.53%, 19.51%, 0.35%, respectively. Deep nets trained by simple back-propagation perform better ...
ACCUWIND - Methods for classification of cup anemometers
DEFF Research Database (Denmark)
Dahlberg, J.-Å.; Friis Pedersen, Troels; Busche, P.
2006-01-01
the errors associated with the use of cup anemometers, and to develop a classification system for quantification of systematic errors of cup anemometers. This classification system has now been implementedin the IEC 61400-12-1 standard on power performance measurements in annex I and J. The classification...... of cup anemometers requires general external climatic operational ranges to be applied for the analysis of systematic errors. A Class A categoryclassification is connected to reasonably flat sites, and another Class B category is connected to complex terrain, General classification indices are the result...... developed in the CLASSCUP projectand earlier. A number of approaches including the use of two cup anemometer models, two methods of torque coefficient measurement, two angular response measurements, and inclusion and exclusion of influence of friction have been implemented in theclassification process...
Iwahashi, J.; Yamazaki, D.; Matsuoka, M.; Thamarux, P.; Herrick, J.; Yong, A.; Mital, U.
2017-12-01
A seamless model of landform classifications with regional accuracy will be a powerful platform for geophysical studies that forecast geologic hazards. Spatial variability as a function of landform on a global scale was captured in the automated classifications of Iwahashi and Pike (2007) and additional developments are presented here that incorporate more accurate depictions using higher-resolution elevation data than the original 1-km scale Shuttle Radar Topography Mission digital elevation model (DEM). We create polygon-based terrain classifications globally by using the 280-m DEM interpolated from the Multi-Error-Removed Improved-Terrain DEM (MERIT; Yamazaki et al., 2017). The multi-scale pixel-image analysis method, known as Multi-resolution Segmentation (Baatz and Schäpe, 2000), is first used to classify the terrains based on geometric signatures (slope and local convexity) calculated from the 280-m DEM. Next, we apply the machine learning method of "k-means clustering" to prepare the polygon-based classification at the globe-scale using slope, local convexity and surface texture. We then group the divisions with similar properties by hierarchical clustering and other statistical analyses using geological and geomorphological data of the area where landslides and earthquakes are frequent (e.g. Japan and California). We find the 280-m DEM resolution is only partially sufficient for classifying plains. We nevertheless observe that the categories correspond to reported landslide and liquefaction features at the global scale, suggesting that our model is an appropriate platform to forecast ground failure. To predict seismic amplification, we estimate site conditions using the time-averaged shear-wave velocity in the upper 30-m (VS30) measurements compiled by Yong et al. (2016) and the terrain model developed by Yong (2016; Y16). We plan to test our method on finer resolution DEMs and report our findings to obtain a more globally consistent terrain model as there
Rueckauer, Bodo; Lungu, Iulia-Alexandra; Hu, Yuhuang; Pfeiffer, Michael; Liu, Shih-Chii
2017-01-01
Spiking neural networks (SNNs) can potentially offer an efficient way of doing inference because the neurons in the networks are sparsely activated and computations are event-driven. Previous work showed that simple continuous-valued deep Convolutional Neural Networks (CNNs) can be converted into accurate spiking equivalents. These networks did not include certain common operations such as max-pooling, softmax, batch-normalization and Inception-modules. This paper presents spiking equivalents of these operations therefore allowing conversion of nearly arbitrary CNN architectures. We show conversion of popular CNN architectures, including VGG-16 and Inception-v3, into SNNs that produce the best results reported to date on MNIST, CIFAR-10 and the challenging ImageNet dataset. SNNs can trade off classification error rate against the number of available operations whereas deep continuous-valued neural networks require a fixed number of operations to achieve their classification error rate. From the examples of LeNet for MNIST and BinaryNet for CIFAR-10, we show that with an increase in error rate of a few percentage points, the SNNs can achieve more than 2x reductions in operations compared to the original CNNs. This highlights the potential of SNNs in particular when deployed on power-efficient neuromorphic spiking neuron chips, for use in embedded applications.
Directory of Open Access Journals (Sweden)
Bodo Rueckauer
2017-12-01
Full Text Available Spiking neural networks (SNNs can potentially offer an efficient way of doing inference because the neurons in the networks are sparsely activated and computations are event-driven. Previous work showed that simple continuous-valued deep Convolutional Neural Networks (CNNs can be converted into accurate spiking equivalents. These networks did not include certain common operations such as max-pooling, softmax, batch-normalization and Inception-modules. This paper presents spiking equivalents of these operations therefore allowing conversion of nearly arbitrary CNN architectures. We show conversion of popular CNN architectures, including VGG-16 and Inception-v3, into SNNs that produce the best results reported to date on MNIST, CIFAR-10 and the challenging ImageNet dataset. SNNs can trade off classification error rate against the number of available operations whereas deep continuous-valued neural networks require a fixed number of operations to achieve their classification error rate. From the examples of LeNet for MNIST and BinaryNet for CIFAR-10, we show that with an increase in error rate of a few percentage points, the SNNs can achieve more than 2x reductions in operations compared to the original CNNs. This highlights the potential of SNNs in particular when deployed on power-efficient neuromorphic spiking neuron chips, for use in embedded applications.
International Nuclear Information System (INIS)
Kwak, Sang Log; Park, Chan Woo; Shin, Seung Ryoung
2008-08-01
Railway accident analysis results show that accidents cased by human factors are not decreasing, whereas H/W related accidents are steadily decreasing. For the efficient management of human factors, many expertise on design, conditions, safety culture and staffing are required. But current safety management activities on safety critical works are focused on training, due to the limited resource and information. In order to improve railway safety, human factors management requirements for safety critical worker and human error classification is proposed in this report. For this accident analysis, status of safety measure on human factor, safety management system on safety critical worker, current safety planning is analysis
ACCUWIND - Methods for classification of cup anemometers
Energy Technology Data Exchange (ETDEWEB)
Dahlberg, J.Aa.; Friis Pedersen, T.; Busche, P.
2006-05-15
Errors associated with the measurement of wind speed are the major sources of uncertainties in power performance testing of wind turbines. Field comparisons of well-calibrated anemometers show significant and not acceptable difference. The European CLASSCUP project posed the objectives to quantify the errors associated with the use of cup anemometers, and to develop a classification system for quantification of systematic errors of cup anemometers. This classification system has now been implemented in the IEC 61400-12-1 standard on power performance measurements in annex I and J. The classification of cup anemometers requires general external climatic operational ranges to be applied for the analysis of systematic errors. A Class A category classification is connected to reasonably flat sites, and another Class B category is connected to complex terrain, General classification indices are the result of assessment of systematic deviations. The present report focuses on methods that can be applied for assessment of such systematic deviations. A new alternative method for torque coefficient measurements at inclined flow have been developed, which have then been applied and compared to the existing methods developed in the CLASSCUP project and earlier. A number of approaches including the use of two cup anemometer models, two methods of torque coefficient measurement, two angular response measurements, and inclusion and exclusion of influence of friction have been implemented in the classification process in order to assess the robustness of methods. The results of the analysis are presented as classification indices, which are compared and discussed. (au)
Masked and unmasked error-related potentials during continuous control and feedback
Lopes Dias, Catarina; Sburlea, Andreea I.; Müller-Putz, Gernot R.
2018-06-01
The detection of error-related potentials (ErrPs) in tasks with discrete feedback is well established in the brain–computer interface (BCI) field. However, the decoding of ErrPs in tasks with continuous feedback is still in its early stages. Objective. We developed a task in which subjects have continuous control of a cursor’s position by means of a joystick. The cursor’s position was shown to the participants in two different modalities of continuous feedback: normal and jittered. The jittered feedback was created to mimic the instability that could exist if participants controlled the trajectory directly with brain signals. Approach. This paper studies the electroencephalographic (EEG)—measurable signatures caused by a loss of control over the cursor’s trajectory, causing a target miss. Main results. In both feedback modalities, time-locked potentials revealed the typical frontal-central components of error-related potentials. Errors occurring during the jittered feedback (masked errors) were delayed in comparison to errors occurring during normal feedback (unmasked errors). Masked errors displayed lower peak amplitudes than unmasked errors. Time-locked classification analysis allowed a good distinction between correct and error classes (average Cohen-, average TPR = 81.8% and average TNR = 96.4%). Time-locked classification analysis between masked error and unmasked error classes revealed results at chance level (average Cohen-, average TPR = 60.9% and average TNR = 58.3%). Afterwards, we performed asynchronous detection of ErrPs, combining both masked and unmasked trials. The asynchronous detection of ErrPs in a simulated online scenario resulted in an average TNR of 84.0% and in an average TPR of 64.9%. Significance. The time-locked classification results suggest that the masked and unmasked errors were indistinguishable in terms of classification. The asynchronous classification results suggest that the
Development of a classification system for cup anemometers - CLASSCUP
DEFF Research Database (Denmark)
Friis Pedersen, Troels
2003-01-01
the objectives to quantify the errors associated with the use of cup anemometers, and to determine the requirements for an optimum design of a cup anemometer, and to develop a classification system forquantification of systematic errors of cup anemometers. The present report describes this proposed...... classification system. A classification method for cup anemometers has been developed, which proposes general external operational ranges to be used. Anormal category range connected to ideal sites of the IEC power performance standard was made, and another extended category range for complex terrain...... was proposed. General classification indices were proposed for all types of cup anemometers. As a resultof the classification, the cup anemometer will be assigned to a certain class: 0.5, 1, 2, 3 or 5 with corresponding intrinsic errors (%) as a vector instrument (3D) or as a horizontal instrument (2D...
A New Classification Approach Based on Multiple Classification Rules
Zhongmei Zhou
2014-01-01
A good classifier can correctly predict new data for which the class label is unknown, so it is important to construct a high accuracy classifier. Hence, classification techniques are much useful in ubiquitous computing. Associative classification achieves higher classification accuracy than some traditional rule-based classification approaches. However, the approach also has two major deficiencies. First, it generates a very large number of association classification rules, especially when t...
Error calculations statistics in radioactive measurements
International Nuclear Information System (INIS)
Verdera, Silvia
1994-01-01
Basic approach and procedures frequently used in the practice of radioactive measurements.Statistical principles applied are part of Good radiopharmaceutical Practices and quality assurance.Concept of error, classification as systematic and random errors.Statistic fundamentals,probability theories, populations distributions, Bernoulli, Poisson,Gauss, t-test distribution,Ξ2 test, error propagation based on analysis of variance.Bibliography.z table,t-test table, Poisson index ,Ξ2 test
Automated Error Detection in Physiotherapy Training.
Jovanović, Marko; Seiffarth, Johannes; Kutafina, Ekaterina; Jonas, Stephan M
2018-01-01
Manual skills teaching, such as physiotherapy education, requires immediate teacher feedback for the students during the learning process, which to date can only be performed by expert trainers. A machine-learning system trained only on correct performances to classify and score performed movements, to identify sources of errors in the movement and give feedback to the learner. We acquire IMU and sEMG sensor data from a commercial-grade wearable device and construct an HMM-based model for gesture classification, scoring and feedback giving. We evaluate the model on publicly available and self-generated data of an exemplary movement pattern executions. The model achieves an overall accuracy of 90.71% on the public dataset and 98.9% on our dataset. An AUC of 0.99 for the ROC of the scoring method could be achieved to discriminate between correct and untrained incorrect executions. The proposed system demonstrated its suitability for scoring and feedback in manual skills training.
Characteristics of medication errors with parenteral cytotoxic drugs
Fyhr, A; Akselsson, R
2012-01-01
Errors involving cytotoxic drugs have the potential of being fatal and should therefore be prevented. The objective of this article is to identify the characteristics of medication errors involving parenteral cytotoxic drugs in Sweden. A total of 60 cases reported to the national error reporting systems from 1996 to 2008 were reviewed. Classification was made to identify cytotoxic drugs involved, type of error, where the error occurred, error detection mechanism, and consequences for the pati...
Robust Transmission of H.264/AVC Streams Using Adaptive Group Slicing and Unequal Error Protection
Thomos, Nikolaos; Argyropoulos, Savvas; Boulgouris, Nikolaos V.; Strintzis, Michael G.
2006-12-01
We present a novel scheme for the transmission of H.264/AVC video streams over lossy packet networks. The proposed scheme exploits the error-resilient features of H.264/AVC codec and employs Reed-Solomon codes to protect effectively the streams. A novel technique for adaptive classification of macroblocks into three slice groups is also proposed. The optimal classification of macroblocks and the optimal channel rate allocation are achieved by iterating two interdependent steps. Dynamic programming techniques are used for the channel rate allocation process in order to reduce complexity. Simulations clearly demonstrate the superiority of the proposed method over other recent algorithms for transmission of H.264/AVC streams.
Selection of anchor values for human error probability estimation
International Nuclear Information System (INIS)
Buffardi, L.C.; Fleishman, E.A.; Allen, J.A.
1989-01-01
There is a need for more dependable information to assist in the prediction of human errors in nuclear power environments. The major objective of the current project is to establish guidelines for using error probabilities from other task settings to estimate errors in the nuclear environment. This involves: (1) identifying critical nuclear tasks, (2) discovering similar tasks in non-nuclear environments, (3) finding error data for non-nuclear tasks, and (4) establishing error-rate values for the nuclear tasks based on the non-nuclear data. A key feature is the application of a classification system to nuclear and non-nuclear tasks to evaluate their similarities and differences in order to provide a basis for generalizing human error estimates across tasks. During the first eight months of the project, several classification systems have been applied to a sample of nuclear tasks. They are discussed in terms of their potential for establishing task equivalence and transferability of human error rates across situations
Muroi, Maki; Shen, Jay J; Angosta, Alona
2017-02-01
Registered nurses (RNs) play an important role in safe medication administration and patient safety. This study examined a total of 1276 medication error (ME) incident reports made by RNs in hospital inpatient settings in the southwestern region of the United States. The most common drug class associated with MEs was cardiovascular drugs (24.7%). Among this class, anticoagulants had the most errors (11.3%). The antimicrobials was the second most common drug class associated with errors (19.1%) and vancomycin was the most common antimicrobial that caused errors in this category (6.1%). MEs occurred more frequently in the medical-surgical and intensive care units than any other hospital units. Ten percent of MEs reached the patients with harm and 11% reached the patients with increased monitoring. Understanding the contributing factors related to MEs, addressing and eliminating risk of errors across hospital units, and providing education and resources for nurses may help reduce MEs. Copyright © 2016 Elsevier Inc. All rights reserved.
Convolutional neural network with transfer learning for rice type classification
Patel, Vaibhav Amit; Joshi, Manjunath V.
2018-04-01
Presently, rice type is identified manually by humans, which is time consuming and error prone. Therefore, there is a need to do this by machine which makes it faster with greater accuracy. This paper proposes a deep learning based method for classification of rice types. We propose two methods to classify the rice types. In the first method, we train a deep convolutional neural network (CNN) using the given segmented rice images. In the second method, we train a combination of a pretrained VGG16 network and the proposed method, while using transfer learning in which the weights of a pretrained network are used to achieve better accuracy. Our approach can also be used for classification of rice grain as broken or fine. We train a 5-class model for classifying rice types using 4000 training images and another 2- class model for the classification of broken and normal rice using 1600 training images. We observe that despite having distinct rice images, our architecture, pretrained on ImageNet data boosts classification accuracy significantly.
Occupancy classification of position weight matrix-inferred transcription factor binding sites.
Directory of Open Access Journals (Sweden)
Hollis Wright
Full Text Available BACKGROUND: Computational prediction of Transcription Factor Binding Sites (TFBS from sequence data alone is difficult and error-prone. Machine learning techniques utilizing additional environmental information about a predicted binding site (such as distances from the site to particular chromatin features to determine its occupancy/functionality class show promise as methods to achieve more accurate prediction of true TFBS in silico. We evaluate the Bayesian Network (BN and Support Vector Machine (SVM machine learning techniques on four distinct TFBS data sets and analyze their performance. We describe the features that are most useful for classification and contrast and compare these feature sets between the factors. RESULTS: Our results demonstrate good performance of classifiers both on TFBS for transcription factors used for initial training and for TFBS for other factors in cross-classification experiments. We find that distances to chromatin modifications (specifically, histone modification islands as well as distances between such modifications to be effective predictors of TFBS occupancy, though the impact of individual predictors is largely TF specific. In our experiments, Bayesian network classifiers outperform SVM classifiers. CONCLUSIONS: Our results demonstrate good performance of machine learning techniques on the problem of occupancy classification, and demonstrate that effective classification can be achieved using distances to chromatin features. We additionally demonstrate that cross-classification of TFBS is possible, suggesting the possibility of constructing a generalizable occupancy classifier capable of handling TFBS for many different transcription factors.
Achieving minimum-error discrimination of an arbitrary set of laser-light pulses
da Silva, Marcus P.; Guha, Saikat; Dutton, Zachary
2013-05-01
Laser light is widely used for communication and sensing applications, so the optimal discrimination of coherent states—the quantum states of light emitted by an ideal laser—has immense practical importance. Due to fundamental limits imposed by quantum mechanics, such discrimination has a finite minimum probability of error. While concrete optical circuits for the optimal discrimination between two coherent states are well known, the generalization to larger sets of coherent states has been challenging. In this paper, we show how to achieve optimal discrimination of any set of coherent states using a resource-efficient quantum computer. Our construction leverages a recent result on discriminating multicopy quantum hypotheses [Blume-Kohout, Croke, and Zwolak, arXiv:1201.6625]. As illustrative examples, we analyze the performance of discriminating a ternary alphabet and show how the quantum circuit of a receiver designed to discriminate a binary alphabet can be reused in discriminating multimode hypotheses. Finally, we show that our result can be used to achieve the quantum limit on the rate of classical information transmission on a lossy optical channel, which is known to exceed the Shannon rate of all conventional optical receivers.
Classification of brain tumors by means of proton nuclear magnetic resonance (NMR) spectroscopy
International Nuclear Information System (INIS)
Sottile, V.S.; Zanchi, D.E.
2017-01-01
In the present work, at the request of health professionals, a computer application named “ViDa” was developed. The aim of this study is to differentiate brain lesions according to whether or not they are tumors, and their subsequent classification into different tumor types using magnetic resonance spectroscopy (SVS) with an echo time of 30 milliseconds. For this development, different areas of knowledge were integrated, among which are Artificial intelligence, physics, programming, physiopathology, images in medicine, among others. Biomedical imaging can be divided into two stages: the pre-processing, performed by the resonator, and post-processing software, performed by ViDa, for the interpretation of the data. This application is included within the Medical Informatics area, as it provides assistance for clinical decision making. The role of the biomedical engineer is fulfilled by developing a health technology in response to a manifested real-life problem. The tool developed shows promising results achieving a 100% Sensitivity, 73% Specificity, 77% Positive Predictive Value and 100% Negative Predictive Value reported in 21 cases tested. The correct classifications of the tumor’s origin reach 70%, the classification of non-astrocytic lesions achieves 67% of correct classifications in that the gradation of astrocytomas achieves a 57% of gradations that agree with biopsies and 43% of slight errors. It was possible to develop an application of assistance to the diagnosis, which together with others medical tests, will make it possible to sharpen the diagnoses of brain tumors. (authors) [es
Exploring diversity in ensemble classification: Applications in large area land cover mapping
Mellor, Andrew; Boukir, Samia
2017-07-01
Ensemble classifiers, such as random forests, are now commonly applied in the field of remote sensing, and have been shown to perform better than single classifier systems, resulting in reduced generalisation error. Diversity across the members of ensemble classifiers is known to have a strong influence on classification performance - whereby classifier errors are uncorrelated and more uniformly distributed across ensemble members. The relationship between ensemble diversity and classification performance has not yet been fully explored in the fields of information science and machine learning and has never been examined in the field of remote sensing. This study is a novel exploration of ensemble diversity and its link to classification performance, applied to a multi-class canopy cover classification problem using random forests and multisource remote sensing and ancillary GIS data, across seven million hectares of diverse dry-sclerophyll dominated public forests in Victoria Australia. A particular emphasis is placed on analysing the relationship between ensemble diversity and ensemble margin - two key concepts in ensemble learning. The main novelty of our work is on boosting diversity by emphasizing the contribution of lower margin instances used in the learning process. Exploring the influence of tree pruning on diversity is also a new empirical analysis that contributes to a better understanding of ensemble performance. Results reveal insights into the trade-off between ensemble classification accuracy and diversity, and through the ensemble margin, demonstrate how inducing diversity by targeting lower margin training samples is a means of achieving better classifier performance for more difficult or rarer classes and reducing information redundancy in classification problems. Our findings inform strategies for collecting training data and designing and parameterising ensemble classifiers, such as random forests. This is particularly important in large area
Yan, Wen-juan; Yang, Ming; He, Guo-quan; Qin, Lin; Li, Gang
2014-11-01
In order to identify the diabetic patients by using tongue near-infrared (NIR) spectrum - a spectral classification model of the NIR reflectivity of the tongue tip is proposed, based on the partial least square (PLS) method. 39sample data of tongue tip's NIR spectra are harvested from healthy people and diabetic patients , respectively. After pretreatment of the reflectivity, the spectral data are set as the independent variable matrix, and information of classification as the dependent variables matrix, Samples were divided into two groups - i.e. 53 samples as calibration set and 25 as prediction set - then the PLS is used to build the classification model The constructed modelfrom the 53 samples has the correlation of 0.9614 and the root mean square error of cross-validation (RMSECV) of 0.1387.The predictions for the 25 samples have the correlation of 0.9146 and the RMSECV of 0.2122.The experimental result shows that the PLS method can achieve good classification on features of healthy people and diabetic patients.
The relationship between twelve-month home stimulation and school achievement.
van Doorninck, W J; Caldwell, B M; Wright, C; Frankenburg, W K
1981-09-01
Home Observation for Measurement of the Environment (HOME) was designed to reflect parental support of early cognitive and socioemotional development. 12-month HOME scores were correlated with elementary school achievement, 5--9 years later. 50 low-income children were rank ordered by a weighted average of centile estimates of achievement test scores, letter grades, and curriculum levels in reading and math. 24 children were classified as having significant school achievement problems. The HOME total score correlated significantly, r = .37, with school centile scores among the low-income families. The statistically more appropriate contingency table analysis revealed a 68% correct classification rate and a significantly reduced error rate over random or blanket prediction. The results supported the predictive value of the 12-month HOME for school achievement among low-income families. In an additional sample of 21 middle-income families, there was insufficient variability among HOME scores to allow prediction. The HOME total scores were highly correlated, r = .86, among siblings tested at least 10 months apart.
Institute of Scientific and Technical Information of China (English)
无
2007-01-01
The multifocal electroretinogram (mfERG) is a newly developed electrophysiological technique. In this paper, a classification method is proposed for early diagnosis of the diabetic retinopathy using mfERG data. MfERG records were obtained from eyes of healthy individuals and patients with diabetes at different stages. For each mfERG record, 103 local responses were extracted. Amplitude value of each point on all the mfERG local responses was looked as one potential feature to classify the experimental subjects. Feature subsets were selected from the feature space by comparing the inter-intra distance. Based on the selected feature subset, Fisher's linear classifiers were trained. And the final classification decision of the record was made by voting all the classifiers' outputs. Applying the method to classify all experimental subjects, very low error rates were achieved. Some crucial properties of the diabetic retinopathy classification method are also discussed.
Effects of uncertainty and variability on population declines and IUCN Red List classifications.
Rueda-Cediel, Pamela; Anderson, Kurt E; Regan, Tracey J; Regan, Helen M
2018-01-22
The International Union for Conservation of Nature (IUCN) Red List Categories and Criteria is a quantitative framework for classifying species according to extinction risk. Population models may be used to estimate extinction risk or population declines. Uncertainty and variability arise in threat classifications through measurement and process error in empirical data and uncertainty in the models used to estimate extinction risk and population declines. Furthermore, species traits are known to affect extinction risk. We investigated the effects of measurement and process error, model type, population growth rate, and age at first reproduction on the reliability of risk classifications based on projected population declines on IUCN Red List classifications. We used an age-structured population model to simulate true population trajectories with different growth rates, reproductive ages and levels of variation, and subjected them to measurement error. We evaluated the ability of scalar and matrix models parameterized with these simulated time series to accurately capture the IUCN Red List classification generated with true population declines. Under all levels of measurement error tested and low process error, classifications were reasonably accurate; scalar and matrix models yielded roughly the same rate of misclassifications, but the distribution of errors differed; matrix models led to greater overestimation of extinction risk than underestimations; process error tended to contribute to misclassifications to a greater extent than measurement error; and more misclassifications occurred for fast, rather than slow, life histories. These results indicate that classifications of highly threatened taxa (i.e., taxa with low growth rates) under criterion A are more likely to be reliable than for less threatened taxa when assessed with population models. Greater scrutiny needs to be placed on data used to parameterize population models for species with high growth rates
Errors in clinical laboratories or errors in laboratory medicine?
Plebani, Mario
2006-01-01
Laboratory testing is a highly complex process and, although laboratory services are relatively safe, they are not as safe as they could or should be. Clinical laboratories have long focused their attention on quality control methods and quality assessment programs dealing with analytical aspects of testing. However, a growing body of evidence accumulated in recent decades demonstrates that quality in clinical laboratories cannot be assured by merely focusing on purely analytical aspects. The more recent surveys on errors in laboratory medicine conclude that in the delivery of laboratory testing, mistakes occur more frequently before (pre-analytical) and after (post-analytical) the test has been performed. Most errors are due to pre-analytical factors (46-68.2% of total errors), while a high error rate (18.5-47% of total errors) has also been found in the post-analytical phase. Errors due to analytical problems have been significantly reduced over time, but there is evidence that, particularly for immunoassays, interference may have a serious impact on patients. A description of the most frequent and risky pre-, intra- and post-analytical errors and advice on practical steps for measuring and reducing the risk of errors is therefore given in the present paper. Many mistakes in the Total Testing Process are called "laboratory errors", although these may be due to poor communication, action taken by others involved in the testing process (e.g., physicians, nurses and phlebotomists), or poorly designed processes, all of which are beyond the laboratory's control. Likewise, there is evidence that laboratory information is only partially utilized. A recent document from the International Organization for Standardization (ISO) recommends a new, broader definition of the term "laboratory error" and a classification of errors according to different criteria. In a modern approach to total quality, centered on patients' needs and satisfaction, the risk of errors and mistakes
Fisher classifier and its probability of error estimation
Chittineni, C. B.
1979-01-01
Computationally efficient expressions are derived for estimating the probability of error using the leave-one-out method. The optimal threshold for the classification of patterns projected onto Fisher's direction is derived. A simple generalization of the Fisher classifier to multiple classes is presented. Computational expressions are developed for estimating the probability of error of the multiclass Fisher classifier.
Fuzzy One-Class Classification Model Using Contamination Neighborhoods
Directory of Open Access Journals (Sweden)
Lev V. Utkin
2012-01-01
Full Text Available A fuzzy classification model is studied in the paper. It is based on the contaminated (robust model which produces fuzzy expected risk measures characterizing classification errors. Optimal classification parameters of the models are derived by minimizing the fuzzy expected risk. It is shown that an algorithm for computing the classification parameters is reduced to a set of standard support vector machine tasks with weighted data points. Experimental results with synthetic data illustrate the proposed fuzzy model.
Errors in practical measurement in surveying, engineering, and technology
International Nuclear Information System (INIS)
Barry, B.A.; Morris, M.D.
1991-01-01
This book discusses statistical measurement, error theory, and statistical error analysis. The topics of the book include an introduction to measurement, measurement errors, the reliability of measurements, probability theory of errors, measures of reliability, reliability of repeated measurements, propagation of errors in computing, errors and weights, practical application of the theory of errors in measurement, two-dimensional errors and includes a bibliography. Appendices are included which address significant figures in measurement, basic concepts of probability and the normal probability curve, writing a sample specification for a procedure, classification, standards of accuracy, and general specifications of geodetic control surveys, the geoid, the frequency distribution curve and the computer and calculator solution of problems
Evaluation of normalization methods for cDNA microarray data by k-NN classification
Energy Technology Data Exchange (ETDEWEB)
Wu, Wei; Xing, Eric P; Myers, Connie; Mian, Saira; Bissell, Mina J
2004-12-17
Non-biological factors give rise to unwanted variations in cDNA microarray data. There are many normalization methods designed to remove such variations. However, to date there have been few published systematic evaluations of these techniques for removing variations arising from dye biases in the context of downstream, higher-order analytical tasks such as classification. Ten location normalization methods that adjust spatial- and/or intensity-dependent dye biases, and three scale methods that adjust scale differences were applied, individually and in combination, to five distinct, published, cancer biology-related cDNA microarray data sets. Leave-one-out cross-validation (LOOCV) classification error was employed as the quantitative end-point for assessing the effectiveness of a normalization method. In particular, a known classifier, k-nearest neighbor (k-NN), was estimated from data normalized using a given technique, and the LOOCV error rate of the ensuing model was computed. We found that k-NN classifiers are sensitive to dye biases in the data. Using NONRM and GMEDIAN as baseline methods, our results show that single-bias-removal techniques which remove either spatial-dependent dye bias (referred later as spatial effect) or intensity-dependent dye bias (referred later as intensity effect) moderately reduce LOOCV classification errors; whereas double-bias-removal techniques which remove both spatial- and intensity effect reduce LOOCV classification errors even further. Of the 41 different strategies examined, three two-step processes, IGLOESS-SLFILTERW7, ISTSPLINE-SLLOESS and IGLOESS-SLLOESS, all of which removed intensity effect globally and spatial effect locally, appear to reduce LOOCV classification errors most consistently and effectively across all data sets. We also found that the investigated scale normalization methods do not reduce LOOCV classification error. Using LOOCV error of k-NNs as the evaluation criterion, three double
Dual Numbers Approach in Multiaxis Machines Error Modeling
Directory of Open Access Journals (Sweden)
Jaroslav Hrdina
2014-01-01
Full Text Available Multiaxis machines error modeling is set in the context of modern differential geometry and linear algebra. We apply special classes of matrices over dual numbers and propose a generalization of such concept by means of general Weil algebras. We show that the classification of the geometric errors follows directly from the algebraic properties of the matrices over dual numbers and thus the calculus over the dual numbers is the proper tool for the methodology of multiaxis machines error modeling.
Game Design Principles based on Human Error
Directory of Open Access Journals (Sweden)
Guilherme Zaffari
2016-03-01
Full Text Available This paper displays the result of the authors’ research regarding to the incorporation of Human Error, through design principles, to video game design. In a general way, designers must consider Human Error factors throughout video game interface development; however, when related to its core design, adaptations are in need, since challenge is an important factor for fun and under the perspective of Human Error, challenge can be considered as a flaw in the system. The research utilized Human Error classifications, data triangulation via predictive human error analysis, and the expanded flow theory to allow the design of a set of principles in order to match the design of playful challenges with the principles of Human Error. From the results, it was possible to conclude that the application of Human Error in game design has a positive effect on player experience, allowing it to interact only with errors associated with the intended aesthetics of the game.
Classification system for reporting events involving human malfunctions
International Nuclear Information System (INIS)
Rasmussen, J.; Pedersen, O.M.; Mancini, G.
1981-01-01
The report describes a set of categories for reporting industrial incidents and events involving human malfunction. The classification system aims at ensuring information adequate for improvement of human work situations and man-machine interface systems and for attempts to quantify ''human error'' rates. The classification system has a multifacetted non-hierarchical structure and its compatibility with Ispra's ERDS classification is described. The collection of the information in general and for quantification purposes are discussed. 24 categories, 12 of which being human factors-oriented, are listed with their respective subcategories, and comments are given. Underlying models of human data process and their typical malfuntions and of a human decision sequence are described. The work reported is a joint contribution to the CSNI Group of Experts on Human Error Data and Assessment
A model-based and computer-aided approach to analysis of human errors in nuclear power plants
International Nuclear Information System (INIS)
Yoon, Wan C.; Lee, Yong H.; Kim, Young S.
1996-01-01
Since the operator's mission in NPPs is increasingly defined by cognitive tasks such as monitoring, diagnosis and planning, the focus of human error analysis should also move from external actions to internal decision-making processes. While more elaborate analysis of cognitive aspects of human errors will help understand their causes and derive effective countermeasures, a lack of framework and an arbitrary resolution of description may hamper the effectiveness of such analysis. This paper presents new model-based schemes of event description and error classification as well as an interactive computerized support system. The schemes and the support system were produced in an effort to develop an improved version of HPES. The use of a decision-making model enables the analyst to document cognitive aspects of human performance explicitly and in a proper resolution. The stage-specific terms used in the proposed schemes make the task of characterizing human errors easier and confident for field analysts. The support system was designed to help the analyst achieve a contextually well-integrated analysis throughout the different parts of HPES
OmniGA: Optimized Omnivariate Decision Trees for Generalizable Classification Models
Magana-Mora, Arturo
2017-06-14
Classification problems from different domains vary in complexity, size, and imbalance of the number of samples from different classes. Although several classification models have been proposed, selecting the right model and parameters for a given classification task to achieve good performance is not trivial. Therefore, there is a constant interest in developing novel robust and efficient models suitable for a great variety of data. Here, we propose OmniGA, a framework for the optimization of omnivariate decision trees based on a parallel genetic algorithm, coupled with deep learning structure and ensemble learning methods. The performance of the OmniGA framework is evaluated on 12 different datasets taken mainly from biomedical problems and compared with the results obtained by several robust and commonly used machine-learning models with optimized parameters. The results show that OmniGA systematically outperformed these models for all the considered datasets, reducing the F score error in the range from 100% to 2.25%, compared to the best performing model. This demonstrates that OmniGA produces robust models with improved performance. OmniGA code and datasets are available at www.cbrc.kaust.edu.sa/omniga/.
OmniGA: Optimized Omnivariate Decision Trees for Generalizable Classification Models
Magana-Mora, Arturo; Bajic, Vladimir B.
2017-01-01
Classification problems from different domains vary in complexity, size, and imbalance of the number of samples from different classes. Although several classification models have been proposed, selecting the right model and parameters for a given classification task to achieve good performance is not trivial. Therefore, there is a constant interest in developing novel robust and efficient models suitable for a great variety of data. Here, we propose OmniGA, a framework for the optimization of omnivariate decision trees based on a parallel genetic algorithm, coupled with deep learning structure and ensemble learning methods. The performance of the OmniGA framework is evaluated on 12 different datasets taken mainly from biomedical problems and compared with the results obtained by several robust and commonly used machine-learning models with optimized parameters. The results show that OmniGA systematically outperformed these models for all the considered datasets, reducing the F score error in the range from 100% to 2.25%, compared to the best performing model. This demonstrates that OmniGA produces robust models with improved performance. OmniGA code and datasets are available at www.cbrc.kaust.edu.sa/omniga/.
A Comparative Study on Error Analysis
DEFF Research Database (Denmark)
Wu, Xiaoli; Zhang, Chun
2015-01-01
Title: A Comparative Study on Error Analysis Subtitle: - Belgian (L1) and Danish (L1) learners’ use of Chinese (L2) comparative sentences in written production Xiaoli Wu, Chun Zhang Abstract: Making errors is an inevitable and necessary part of learning. The collection, classification and analysis...... the occurrence of errors either in linguistic or pedagogical terms. The purpose of the current study is to demonstrate the theoretical and practical relevance of error analysis approach in CFL by investigating two cases - (1) Belgian (L1) learners’ use of Chinese (L2) comparative sentences in written production...... of errors in the written and spoken production of L2 learners has a long tradition in L2 pedagogy. Yet, in teaching and learning Chinese as a foreign language (CFL), only handful studies have been made either to define the ‘error’ in a pedagogically insightful way or to empirically investigate...
Xie, W.-J.; Zhang, L.; Chen, H.-P.; Zhou, J.; Mao, W.-J.
2018-04-01
The purpose of carrying out national geographic conditions monitoring is to obtain information of surface changes caused by human social and economic activities, so that the geographic information can be used to offer better services for the government, enterprise and public. Land cover data contains detailed geographic conditions information, thus has been listed as one of the important achievements in the national geographic conditions monitoring project. At present, the main issue of the production of the land cover data is about how to improve the classification accuracy. For the land cover data quality inspection and acceptance, classification accuracy is also an important check point. So far, the classification accuracy inspection is mainly based on human-computer interaction or manual inspection in the project, which are time consuming and laborious. By harnessing the automatic high-resolution remote sensing image change detection technology based on the ERDAS IMAGINE platform, this paper carried out the classification accuracy inspection test of land cover data in the project, and presented a corresponding technical route, which includes data pre-processing, change detection, result output and information extraction. The result of the quality inspection test shows the effectiveness of the technical route, which can meet the inspection needs for the two typical errors, that is, missing and incorrect update error, and effectively reduces the work intensity of human-computer interaction inspection for quality inspectors, and also provides a technical reference for the data production and quality control of the land cover data.
A Confidence Paradigm for Classification Systems
2008-09-01
methodology to determine how much confi- dence one should have in a classifier output. This research proposes a framework to determine the level of...theoretical framework that attempts to unite the viewpoints of the classification system developer (or engineer) and the classification system user (or...operating point. An algorithm is developed that minimizes a “confidence” measure called Binned Error in the Posterior ( BEP ). Then, we prove that training a
Can Automatic Classification Help to Increase Accuracy in Data Collection?
Directory of Open Access Journals (Sweden)
Frederique Lang
2016-09-01
classification achieved by this means is not completely accurate, the amount of manual coding needed can be greatly reduced by using classification algorithms. This can be of great help when the dataset is big. With the help of accuracy, recall, and coverage measures, it is possible to have an estimation of the error involved in this classification, which could open the possibility of incorporating the use of these algorithms in software specifically designed for data cleaning and classification. Originality/value: We analyzed the performance of seven algorithms and whether combinations of these algorithms improve accuracy in data collection. Use of these algorithms could reduce time needed for manual data cleaning.
Land-cover classification with an expert classification algorithm using digital aerial photographs
Directory of Open Access Journals (Sweden)
José L. de la Cruz
2010-05-01
Full Text Available The purpose of this study was to evaluate the usefulness of the spectral information of digital aerial sensors in determining land-cover classification using new digital techniques. The land covers that have been evaluated are the following, (1 bare soil, (2 cereals, including maize (Zea mays L., oats (Avena sativa L., rye (Secale cereale L., wheat (Triticum aestivum L. and barley (Hordeun vulgare L., (3 high protein crops, such as peas (Pisum sativum L. and beans (Vicia faba L., (4 alfalfa (Medicago sativa L., (5 woodlands and scrublands, including holly oak (Quercus ilex L. and common retama (Retama sphaerocarpa L., (6 urban soil, (7 olive groves (Olea europaea L. and (8 burnt crop stubble. The best result was obtained using an expert classification algorithm, achieving a reliability rate of 95%. This result showed that the images of digital airborne sensors hold considerable promise for the future in the field of digital classifications because these images contain valuable information that takes advantage of the geometric viewpoint. Moreover, new classification techniques reduce problems encountered using high-resolution images; while reliabilities are achieved that are better than those achieved with traditional methods.
Spectrum of diagnostic errors in radiology.
Pinto, Antonio; Brunese, Luca
2010-10-28
Diagnostic errors are important in all branches of medicine because they are an indication of poor patient care. Since the early 1970s, physicians have been subjected to an increasing number of medical malpractice claims. Radiology is one of the specialties most liable to claims of medical negligence. Most often, a plaintiff's complaint against a radiologist will focus on a failure to diagnose. The etiology of radiological error is multi-factorial. Errors fall into recurrent patterns. Errors arise from poor technique, failures of perception, lack of knowledge and misjudgments. The work of diagnostic radiology consists of the complete detection of all abnormalities in an imaging examination and their accurate diagnosis. Every radiologist should understand the sources of error in diagnostic radiology as well as the elements of negligence that form the basis of malpractice litigation. Error traps need to be uncovered and highlighted, in order to prevent repetition of the same mistakes. This article focuses on the spectrum of diagnostic errors in radiology, including a classification of the errors, and stresses the malpractice issues in mammography, chest radiology and obstetric sonography. Missed fractures in emergency and communication issues between radiologists and physicians are also discussed.
Automatic Classification of Aerial Imagery for Urban Hydrological Applications
Paul, A.; Yang, C.; Breitkopf, U.; Liu, Y.; Wang, Z.; Rottensteiner, F.; Wallner, M.; Verworn, A.; Heipke, C.
2018-04-01
In this paper we investigate the potential of automatic supervised classification for urban hydrological applications. In particular, we contribute to runoff simulations using hydrodynamic urban drainage models. In order to assess whether the capacity of the sewers is sufficient to avoid surcharge within certain return periods, precipitation is transformed into runoff. The transformation of precipitation into runoff requires knowledge about the proportion of drainage-effective areas and their spatial distribution in the catchment area. Common simulation methods use the coefficient of imperviousness as an important parameter to estimate the overland flow, which subsequently contributes to the pipe flow. The coefficient of imperviousness is the percentage of area covered by impervious surfaces such as roofs or road surfaces. It is still common practice to assign the coefficient of imperviousness for each particular land parcel manually by visual interpretation of aerial images. Based on classification results of these imagery we contribute to an objective automatic determination of the coefficient of imperviousness. In this context we compare two classification techniques: Random Forests (RF) and Conditional Random Fields (CRF). Experimental results performed on an urban test area show good results and confirm that the automated derivation of the coefficient of imperviousness, apart from being more objective and, thus, reproducible, delivers more accurate results than the interactive estimation. We achieve an overall accuracy of about 85 % for both classifiers. The root mean square error of the differences of the coefficient of imperviousness compared to the reference is 4.4 % for the CRF-based classification, and 3.8 % for the RF-based classification.
2017-06-06
information. Send comments regarding this burden estimate or any other aspect of this collection of information, including suggestions for reducing this...ms. 4 2.2 CLASSIFICATION Target classification is usually carried out using cued interrogation data acquired over anomalies initially...identified in the detection data. These cued interrogations eliminate relative positional errors by acquiring data with a stationary sensor. The multi-static
A qualitative description of human error
International Nuclear Information System (INIS)
Li Zhaohuan
1992-11-01
The human error has an important contribution to risk of reactor operation. The insight and analytical model are main parts in human reliability analysis. It consists of the concept of human error, the nature, the mechanism of generation, the classification and human performance influence factors. On the operating reactor the human error is defined as the task-human-machine mismatch. The human error event is focused on the erroneous action and the unfavored result. From the time limitation of performing a task, the operation is divided into time-limited and time-opened. The HCR (human cognitive reliability) model is suited for only time-limited. The basic cognitive process consists of the information gathering, cognition/thinking, decision making and action. The human erroneous action may be generated in any stage of this process. The more natural ways to classify human errors are presented. The human performance influence factors including personal, organizational and environmental factors are also listed
A qualitative description of human error
Energy Technology Data Exchange (ETDEWEB)
Zhaohuan, Li [Academia Sinica, Beijing, BJ (China). Inst. of Atomic Energy
1992-11-01
The human error has an important contribution to risk of reactor operation. The insight and analytical model are main parts in human reliability analysis. It consists of the concept of human error, the nature, the mechanism of generation, the classification and human performance influence factors. On the operating reactor the human error is defined as the task-human-machine mismatch. The human error event is focused on the erroneous action and the unfavored result. From the time limitation of performing a task, the operation is divided into time-limited and time-opened. The HCR (human cognitive reliability) model is suited for only time-limited. The basic cognitive process consists of the information gathering, cognition/thinking, decision making and action. The human erroneous action may be generated in any stage of this process. The more natural ways to classify human errors are presented. The human performance influence factors including personal, organizational and environmental factors are also listed.
A Comparative Analysis of Classification Algorithms on Diverse Datasets
Directory of Open Access Journals (Sweden)
M. Alghobiri
2018-04-01
Full Text Available Data mining involves the computational process to find patterns from large data sets. Classification, one of the main domains of data mining, involves known structure generalizing to apply to a new dataset and predict its class. There are various classification algorithms being used to classify various data sets. They are based on different methods such as probability, decision tree, neural network, nearest neighbor, boolean and fuzzy logic, kernel-based etc. In this paper, we apply three diverse classification algorithms on ten datasets. The datasets have been selected based on their size and/or number and nature of attributes. Results have been discussed using some performance evaluation measures like precision, accuracy, F-measure, Kappa statistics, mean absolute error, relative absolute error, ROC Area etc. Comparative analysis has been carried out using the performance evaluation measures of accuracy, precision, and F-measure. We specify features and limitations of the classification algorithms for the diverse nature datasets.
Funk, Kerri L.; Tseng, M. S.
Two groups of 32 educable mentally retarded children (ages 7 to 14 years) were compared as to their arithmetic and classification performances attributable to the presence or absence of a 4 1/2 week exposure to classification tasks. The randomized block pretest-posttest design was used. The experimental group and the control group were matched on…
DOE LLW classification rationale
International Nuclear Information System (INIS)
Flores, A.Y.
1991-01-01
This report was about the rationale which the US Department of Energy had with low-level radioactive waste (LLW) classification. It is based on the Nuclear Regulatory Commission's classification system. DOE site operators met to review the qualifications and characteristics of the classification systems. They evaluated performance objectives, developed waste classification tables, and compiled dose limits on the waste. A goal of the LLW classification system was to allow each disposal site the freedom to develop limits to radionuclide inventories and concentrations according to its own site-specific characteristics. This goal was achieved with the adoption of a performance objectives system based on a performance assessment, with site-specific environmental conditions and engineered disposal systems
Review of human error analysis methodologies and case study for accident management
International Nuclear Information System (INIS)
Jung, Won Dae; Kim, Jae Whan; Lee, Yong Hee; Ha, Jae Joo
1998-03-01
In this research, we tried to establish the requirements for the development of a new human error analysis method. To achieve this goal, we performed a case study as following steps; 1. review of the existing HEA methods 2. selection of those methods which are considered to be appropriate for the analysis of operator's tasks in NPPs 3. choice of tasks for the application, selected for the case study: HRMS (Human reliability management system), PHECA (Potential Human Error Cause Analysis), CREAM (Cognitive Reliability and Error Analysis Method). And, as the tasks for the application, 'bleed and feed operation' and 'decision-making for the reactor cavity flooding' tasks are chosen. We measured the applicability of the selected methods to the NPP tasks, and evaluated the advantages and disadvantages between each method. The three methods are turned out to be applicable for the prediction of human error. We concluded that both of CREAM and HRMS are equipped with enough applicability for the NPP tasks, however, compared two methods. CREAM is thought to be more appropriate than HRMS from the viewpoint of overall requirements. The requirements for the new HEA method obtained from the study can be summarized as follows; firstly, it should deal with cognitive error analysis, secondly, it should have adequate classification system for the NPP tasks, thirdly, the description on the error causes and error mechanisms should be explicit, fourthly, it should maintain the consistency of the result by minimizing the ambiguity in each step of analysis procedure, fifty, it should be done with acceptable human resources. (author). 25 refs., 30 tabs., 4 figs
Automated classification of Acid Rock Drainage potential from Corescan drill core imagery
Cracknell, M. J.; Jackson, L.; Parbhakar-Fox, A.; Savinova, K.
2017-12-01
Classification of the acid forming potential of waste rock is important for managing environmental hazards associated with mining operations. Current methods for the classification of acid rock drainage (ARD) potential usually involve labour intensive and subjective assessment of drill core and/or hand specimens. Manual methods are subject to operator bias, human error and the amount of material that can be assessed within a given time frame is limited. The automated classification of ARD potential documented here is based on the ARD Index developed by Parbhakar-Fox et al. (2011). This ARD Index involves the combination of five indicators: A - sulphide content; B - sulphide alteration; C - sulphide morphology; D - primary neutraliser content; and E - sulphide mineral association. Several components of the ARD Index require accurate identification of sulphide minerals. This is achieved by classifying Corescan Red-Green-Blue true colour images into the presence or absence of sulphide minerals using supervised classification. Subsequently, sulphide classification images are processed and combined with Corescan SWIR-based mineral classifications to obtain information on sulphide content, indices representing sulphide textures (disseminated versus massive and degree of veining), and spatially associated minerals. This information is combined to calculate ARD Index indicator values that feed into the classification of ARD potential. Automated ARD potential classifications of drill core samples associated with a porphyry Cu-Au deposit are compared to manually derived classifications and those obtained by standard static geochemical testing and X-ray diffractometry analyses. Results indicate a high degree of similarity between automated and manual ARD potential classifications. Major differences between approaches are observed in sulphide and neutraliser mineral percentages, likely due to the subjective nature of manual estimates of mineral content. The automated approach
White, Simon R; Muniz-Terrera, Graciela; Matthews, Fiona E
2018-05-01
Many medical (and ecological) processes involve the change of shape, whereby one trajectory changes into another trajectory at a specific time point. There has been little investigation into the study design needed to investigate these models. We consider the class of fixed effect change-point models with an underlying shape comprised two joined linear segments, also known as broken-stick models. We extend this model to include two sub-groups with different trajectories at the change-point, a change and no change class, and also include a missingness model to account for individuals with incomplete follow-up. Through a simulation study, we consider the relationship of sample size to the estimates of the underlying shape, the existence of a change-point, and the classification-error of sub-group labels. We use a Bayesian framework to account for the missing labels, and the analysis of each simulation is performed using standard Markov chain Monte Carlo techniques. Our simulation study is inspired by cognitive decline as measured by the Mini-Mental State Examination, where our extended model is appropriate due to the commonly observed mixture of individuals within studies who do or do not exhibit accelerated decline. We find that even for studies of modest size ( n = 500, with 50 individuals observed past the change-point) in the fixed effect setting, a change-point can be detected and reliably estimated across a range of observation-errors.
Errors in imaging patients in the emergency setting.
Pinto, Antonio; Reginelli, Alfonso; Pinto, Fabio; Lo Re, Giuseppe; Midiri, Federico; Muzj, Carlo; Romano, Luigia; Brunese, Luca
2016-01-01
Emergency and trauma care produces a "perfect storm" for radiological errors: uncooperative patients, inadequate histories, time-critical decisions, concurrent tasks and often junior personnel working after hours in busy emergency departments. The main cause of diagnostic errors in the emergency department is the failure to correctly interpret radiographs, and the majority of diagnoses missed on radiographs are fractures. Missed diagnoses potentially have important consequences for patients, clinicians and radiologists. Radiologists play a pivotal role in the diagnostic assessment of polytrauma patients and of patients with non-traumatic craniothoracoabdominal emergencies, and key elements to reduce errors in the emergency setting are knowledge, experience and the correct application of imaging protocols. This article aims to highlight the definition and classification of errors in radiology, the causes of errors in emergency radiology and the spectrum of diagnostic errors in radiography, ultrasonography and CT in the emergency setting.
Classification of DNA nucleotides with transverse tunneling currents
Pedersen, Jonas Nyvold; Boynton, Paul; Ventra, Massimiliano Di; Jauho, Antti-Pekka; Flyvbjerg, Henrik
2016-01-01
It has been theoretically suggested and experimentally demonstrated that fast and low-cost sequencing of DNA, RNA, and peptide molecules might be achieved by passing such molecules between electrodes embedded in a nanochannel. The experimental realization of this scheme faces major challenges, however. In realistic liquid environments, typical currents in tunnelling devices are of the order of picoamps. This corresponds to only six electrons per microsecond, and this number affects the integration time required to do current measurements in real experiments. This limits the speed of sequencing, though current fluctuations due to Brownian motion of the molecule average out during the required integration time. Moreover, data acquisition equipment introduces noise, and electronic filters create correlations in time-series data. We discuss how these effects must be included in the analysis of, e.g., the assignment of specific nucleobases to current signals. As the signals from different molecules overlap, unambiguous classification is impossible with a single measurement. We argue that the assignment of molecules to a signal is a standard pattern classification problem and calculation of the error rates is straightforward. The ideas presented here can be extended to other sequencing approaches of current interest. PMID:27897144
Classification of DNA nucleotides with transverse tunneling currents
Nyvold Pedersen, Jonas; Boynton, Paul; Di Ventra, Massimiliano; Jauho, Antti-Pekka; Flyvbjerg, Henrik
2017-01-01
It has been theoretically suggested and experimentally demonstrated that fast and low-cost sequencing of DNA, RNA, and peptide molecules might be achieved by passing such molecules between electrodes embedded in a nanochannel. The experimental realization of this scheme faces major challenges, however. In realistic liquid environments, typical currents in tunneling devices are of the order of picoamps. This corresponds to only six electrons per microsecond, and this number affects the integration time required to do current measurements in real experiments. This limits the speed of sequencing, though current fluctuations due to Brownian motion of the molecule average out during the required integration time. Moreover, data acquisition equipment introduces noise, and electronic filters create correlations in time-series data. We discuss how these effects must be included in the analysis of, e.g., the assignment of specific nucleobases to current signals. As the signals from different molecules overlap, unambiguous classification is impossible with a single measurement. We argue that the assignment of molecules to a signal is a standard pattern classification problem and calculation of the error rates is straightforward. The ideas presented here can be extended to other sequencing approaches of current interest.
Unsupervised classification of operator workload from brain signals
Schultze-Kraft, Matthias; Dähne, Sven; Gugler, Manfred; Curio, Gabriel; Blankertz, Benjamin
2016-06-01
Objective. In this study we aimed for the classification of operator workload as it is expected in many real-life workplace environments. We explored brain-signal based workload predictors that differ with respect to the level of label information required for training, including entirely unsupervised approaches. Approach. Subjects executed a task on a touch screen that required continuous effort of visual and motor processing with alternating difficulty. We first employed classical approaches for workload state classification that operate on the sensor space of EEG and compared those to the performance of three state-of-the-art spatial filtering methods: common spatial patterns (CSPs) analysis, which requires binary label information; source power co-modulation (SPoC) analysis, which uses the subjects’ error rate as a target function; and canonical SPoC (cSPoC) analysis, which solely makes use of cross-frequency power correlations induced by different states of workload and thus represents an unsupervised approach. Finally, we investigated the effects of fusing brain signals and peripheral physiological measures (PPMs) and examined the added value for improving classification performance. Main results. Mean classification accuracies of 94%, 92% and 82% were achieved with CSP, SPoC, cSPoC, respectively. These methods outperformed the approaches that did not use spatial filtering and they extracted physiologically plausible components. The performance of the unsupervised cSPoC is significantly increased by augmenting it with PPM features. Significance. Our analyses ensured that the signal sources used for classification were of cortical origin and not contaminated with artifacts. Our findings show that workload states can be successfully differentiated from brain signals, even when less and less information from the experimental paradigm is used, thus paving the way for real-world applications in which label information may be noisy or entirely unavailable.
Directory of Open Access Journals (Sweden)
Rodrigo de Sales
2017-09-01
Full Text Available The studies of library classification generally interact with the historical contextualization approach and with the classification ideas typical of Philosophy. In the 19th century, the North-American philosopher and educator William Torrey Harris developed a book classification at the St. Louis Public School, based on Francis Bacon and Georg Wilhelm Friedrich Hegel. The objective of this essay is to analyze Harris’s classification, reflecting upon his theoretical and philosophical backgrounds. To achieve such objective, this essay adopts a critical-descriptive approach for analysis. Results show some influences of Bacon and Hegel in Harris’s classification.
Software errors and complexity: An empirical investigation
Basili, Victor R.; Perricone, Berry T.
1983-01-01
The distributions and relationships derived from the change data collected during the development of a medium scale satellite software project show that meaningful results can be obtained which allow an insight into software traits and the environment in which it is developed. Modified and new modules were shown to behave similarly. An abstract classification scheme for errors which allows a better understanding of the overall traits of a software project is also shown. Finally, various size and complexity metrics are examined with respect to errors detected within the software yielding some interesting results.
International Nuclear Information System (INIS)
Takagawa, Kenichi
2008-01-01
Human errors by maintenance workers in U.S. nuclear power plants were compared with those in Japanese nuclear power plants for the same period in order to identify the characteristics of such errors. As for U.S. events, cases which occurred during 2006 were selected from the Nuclear Information Database of the Institute to Nuclear Safety System while Japanese cases that occurred during the same period, were extracted from the Nuclear Information Archives (NUCIA) owned by JANTI. The most common cause of human errors was insufficient knowledge or experience' accounting for about 40% for U.S. cases and 50% or more of cases in Japan. To break down 'insufficient knowledge', we classified the contents of knowledge into five categories; method', 'nature', 'reason', 'scope' and 'goal', and classified the level of knowledge into four categories: 'known', 'comprehended', 'applied' and analytic'. By using this classification, the patterns of combination of each item of the content and the level of knowledge were compared. In the U.S. cases, errors due to 'insufficient knowledge of nature and insufficient knowledge of method' were prevalent while three other items', 'reason', scope' and 'goal' which involve work conditions among the contents of knowledge rarely occurred. In Japan, errors arising from 'nature not being comprehended' were rather prevalent while other cases were distributed evenly for all categories including the work conditions. For addressing insufficient knowledge or experience', we consider that the following approaches are valid: according to the knowledge level which is required for the work, the reflection of knowledge on the procedure or education materials, training and confirmation of understanding level, virtual practice and instruction of experience should be implemented. As for the knowledge on the work conditions, it is necessary to enter the work conditions in the procedure and education materials while conducting training or education. (author)
Representation Learning for Class C G Protein-Coupled Receptors Classification
Directory of Open Access Journals (Sweden)
Raúl Cruz-Barbosa
2018-03-01
Full Text Available G protein-coupled receptors (GPCRs are integral cell membrane proteins of relevance for pharmacology. The complete tertiary structure including both extracellular and transmembrane domains has not been determined for any member of class C GPCRs. An alternative way to work on GPCR structural models is the investigation of their functionality through the analysis of their primary structure. For this, sequence representation is a key factor for the GPCRs’ classification context, where usually, feature engineering is carried out. In this paper, we propose the use of representation learning to acquire the features that best represent the class C GPCR sequences and at the same time to obtain a model for classification automatically. Deep learning methods in conjunction with amino acid physicochemical property indices are then used for this purpose. Experimental results assessed by the classification accuracy, Matthews’ correlation coefficient and the balanced error rate show that using a hydrophobicity index and a restricted Boltzmann machine (RBM can achieve performance results (accuracy of 92.9% similar to those reported in the literature. As a second proposal, we combine two or more physicochemical property indices instead of only one as the input for a deep architecture in order to add information from the sequences. Experimental results show that using three hydrophobicity-related index combinations helps to improve the classification performance (accuracy of 94.1% of an RBM better than those reported in the literature for class C GPCRs without using feature selection methods.
Energy Technology Data Exchange (ETDEWEB)
Diego-Vallejo, David [Technische Universität Berlin, Institute of Optics and Atomic Physics, Straße des 17, Juni 135, 10623 Berlin (Germany); Laser- und Medizin- Technologie Berlin GmbH (LMTB), Applied Laser Technology, Fabeckstr. 60-62, 14195 Berlin (Germany); Ashkenasi, David, E-mail: d.ashkenasi@lmtb.de [Laser- und Medizin- Technologie Berlin GmbH (LMTB), Applied Laser Technology, Fabeckstr. 60-62, 14195 Berlin (Germany); Lemke, Andreas [Laser- und Medizin- Technologie Berlin GmbH (LMTB), Applied Laser Technology, Fabeckstr. 60-62, 14195 Berlin (Germany); Eichler, Hans Joachim [Technische Universität Berlin, Institute of Optics and Atomic Physics, Straße des 17, Juni 135, 10623 Berlin (Germany); Laser- und Medizin- Technologie Berlin GmbH (LMTB), Applied Laser Technology, Fabeckstr. 60-62, 14195 Berlin (Germany)
2013-09-01
Laser-induced breakdown spectroscopy (LIBS) and two classification methods, i.e. linear correlation and artificial neural networks (ANN), are used to monitor P1, P2 and P3 scribing steps of Copper-Indium-Diselenide (CIS) solar cells. Narrow channels featuring complete removal of desired layers with minimum damage on the underlying film are expected to enhance efficiency of solar cells. The monitoring technique is intended to determine that enough material has been removed to reach the desired layer based on the analysis of plasma emission acquired during multiple pass laser scribing. When successful selective scribing is achieved, a high degree of similarity between test and reference spectra has to be identified by classification methods in order to stop the scribing procedure and avoid damaging the bottom layer. Performance of linear correlation and artificial neural networks is compared and evaluated for two spectral bandwidths. By using experimentally determined combinations of classifier and analyzed spectral band for each step, classification performance achieves errors of 7, 1 and 4% for steps P1, P2 and P3, respectively. The feasibility of using plasma emission for the supervision of processing steps of solar cell manufacturing is demonstrated. This method has the potential to be implemented as an online monitoring procedure assisting the production of solar cells. - Highlights: • LIBS and two classification methods were used to monitor CIS solar cells processing. • Selective ablation of thin-film solar cells was improved with inspection system. • Customized classification method and analyzed spectral band enhanced performance.
Feature extraction and classification in automatic weld seam radioscopy
International Nuclear Information System (INIS)
Heindoerfer, F.; Pohle, R.
1994-01-01
The investigations conducted have shown that automatic feature extraction and classification procedures permit the identification of weld seam flaws. Within this context the favored learning fuzzy classificator represents a very good alternative to conventional classificators. The results have also made clear that improvements mainly in the field of image registration are still possible by increasing the resolution of the radioscopy system. Since, only if the flaw is segmented correctly, i.e. in its full size, and due to improved detail recognizability and sufficient contrast difference will an almost error-free classification be conceivable. (orig./MM) [de
Classification of quantitative light-induced fluorescence images using convolutional neural network
Imangaliyev, S.; van der Veen, M.H.; Volgenant, C.M.C.; Loos, B.G.; Keijser, B.J.F.; Crielaard, W.; Levin, E.; Lintas, A.; Rovetta, S.; Verschure, P.F.M.J.; Villa, A.E.P.
2017-01-01
Images are an important data source for diagnosis of oral diseases. The manual classification of images may lead to suboptimal treatment procedures due to subjective errors. In this paper an image classification algorithm based on Deep Learning framework is applied to Quantitative Light-induced
Quantum learning: asymptotically optimal classification of qubit states
International Nuclear Information System (INIS)
Guta, Madalin; Kotlowski, Wojciech
2010-01-01
Pattern recognition is a central topic in learning theory, with numerous applications such as voice and text recognition, image analysis and computer diagnosis. The statistical setup in classification is the following: we are given an i.i.d. training set (X 1 , Y 1 ), ... , (X n , Y n ), where X i represents a feature and Y i in{0, 1} is a label attached to that feature. The underlying joint distribution of (X, Y) is unknown, but we can learn about it from the training set, and we aim at devising low error classifiers f: X→Y used to predict the label of new incoming features. In this paper, we solve a quantum analogue of this problem, namely the classification of two arbitrary unknown mixed qubit states. Given a number of 'training' copies from each of the states, we would like to 'learn' about them by performing a measurement on the training set. The outcome is then used to design measurements for the classification of future systems with unknown labels. We found the asymptotically optimal classification strategy and show that typically it performs strictly better than a plug-in strategy, which consists of estimating the states separately and then discriminating between them using the Helstrom measurement. The figure of merit is given by the excess risk equal to the difference between the probability of error and the probability of error of the optimal measurement for known states. We show that the excess risk scales as n -1 and compute the exact constant of the rate.
Classification of Mistakes in Patient Care in a Nigerian Hospital ...
African Journals Online (AJOL)
The study shows that there are wide variations within and between professional health groups in the classification of errors in patient care. The implications of the absence of a classificatory scheme for errors in patient care for service improvement and organisational learning in the hospital environment are discussed.
A dictionary learning approach for human sperm heads classification.
Shaker, Fariba; Monadjemi, S Amirhassan; Alirezaie, Javad; Naghsh-Nilchi, Ahmad Reza
2017-12-01
To diagnose infertility in men, semen analysis is conducted in which sperm morphology is one of the factors that are evaluated. Since manual assessment of sperm morphology is time-consuming and subjective, automatic classification methods are being developed. Automatic classification of sperm heads is a complicated task due to the intra-class differences and inter-class similarities of class objects. In this research, a Dictionary Learning (DL) technique is utilized to construct a dictionary of sperm head shapes. This dictionary is used to classify the sperm heads into four different classes. Square patches are extracted from the sperm head images. Columnized patches from each class of sperm are used to learn class-specific dictionaries. The patches from a test image are reconstructed using each class-specific dictionary and the overall reconstruction error for each class is used to select the best matching class. Average accuracy, precision, recall, and F-score are used to evaluate the classification method. The method is evaluated using two publicly available datasets of human sperm head shapes. The proposed DL based method achieved an average accuracy of 92.2% on the HuSHeM dataset, and an average recall of 62% on the SCIAN-MorphoSpermGS dataset. The results show a significant improvement compared to a previously published shape-feature-based method. We have achieved high-performance results. In addition, our proposed approach offers a more balanced classifier in which all four classes are recognized with high precision and recall. In this paper, we use a Dictionary Learning approach in classifying human sperm heads. It is shown that the Dictionary Learning method is far more effective in classifying human sperm heads than classifiers using shape-based features. Also, a dataset of human sperm head shapes is introduced to facilitate future research. Copyright © 2017 Elsevier Ltd. All rights reserved.
Bread crumb classification using fractal and multifractal features
Baravalle, Rodrigo Guillermo; Delrieux, Claudio Augusto; Gómez, Juan Carlos
2017-01-01
Adequate image descriptors are fundamental in image classification and object recognition. Main requirements for image features are robustness and low dimensionality which would lead to low classification errors in a variety of situations and with a reasonable computational cost. In this context, the identification of materials poses a significant challenge, since typical (geometric and/or differential) feature extraction methods are not robust enough. Texture features based on Fourier or wav...
Taghanaki, Saeid Asgari; Kawahara, Jeremy; Miles, Brandon; Hamarneh, Ghassan
2017-07-01
Feature reduction is an essential stage in computer aided breast cancer diagnosis systems. Multilayer neural networks can be trained to extract relevant features by encoding high-dimensional data into low-dimensional codes. Optimizing traditional auto-encoders works well only if the initial weights are close to a proper solution. They are also trained to only reduce the mean squared reconstruction error (MRE) between the encoder inputs and the decoder outputs, but do not address the classification error. The goal of the current work is to test the hypothesis that extending traditional auto-encoders (which only minimize reconstruction error) to multi-objective optimization for finding Pareto-optimal solutions provides more discriminative features that will improve classification performance when compared to single-objective and other multi-objective approaches (i.e. scalarized and sequential). In this paper, we introduce a novel multi-objective optimization of deep auto-encoder networks, in which the auto-encoder optimizes two objectives: MRE and mean classification error (MCE) for Pareto-optimal solutions, rather than just MRE. These two objectives are optimized simultaneously by a non-dominated sorting genetic algorithm. We tested our method on 949 X-ray mammograms categorized into 12 classes. The results show that the features identified by the proposed algorithm allow a classification accuracy of up to 98.45%, demonstrating favourable accuracy over the results of state-of-the-art methods reported in the literature. We conclude that adding the classification objective to the traditional auto-encoder objective and optimizing for finding Pareto-optimal solutions, using evolutionary multi-objective optimization, results in producing more discriminative features. Copyright © 2017 Elsevier B.V. All rights reserved.
Directory of Open Access Journals (Sweden)
Hesham Mostafa
2017-09-01
Full Text Available Artificial neural networks (ANNs trained using backpropagation are powerful learning architectures that have achieved state-of-the-art performance in various benchmarks. Significant effort has been devoted to developing custom silicon devices to accelerate inference in ANNs. Accelerating the training phase, however, has attracted relatively little attention. In this paper, we describe a hardware-efficient on-line learning technique for feedforward multi-layer ANNs that is based on pipelined backpropagation. Learning is performed in parallel with inference in the forward pass, removing the need for an explicit backward pass and requiring no extra weight lookup. By using binary state variables in the feedforward network and ternary errors in truncated-error backpropagation, the need for any multiplications in the forward and backward passes is removed, and memory requirements for the pipelining are drastically reduced. Further reduction in addition operations owing to the sparsity in the forward neural and backpropagating error signal paths contributes to highly efficient hardware implementation. For proof-of-concept validation, we demonstrate on-line learning of MNIST handwritten digit classification on a Spartan 6 FPGA interfacing with an external 1Gb DDR2 DRAM, that shows small degradation in test error performance compared to an equivalently sized binary ANN trained off-line using standard back-propagation and exact errors. Our results highlight an attractive synergy between pipelined backpropagation and binary-state networks in substantially reducing computation and memory requirements, making pipelined on-line learning practical in deep networks.
Mostafa, Hesham; Pedroni, Bruno; Sheik, Sadique; Cauwenberghs, Gert
2017-01-01
Artificial neural networks (ANNs) trained using backpropagation are powerful learning architectures that have achieved state-of-the-art performance in various benchmarks. Significant effort has been devoted to developing custom silicon devices to accelerate inference in ANNs. Accelerating the training phase, however, has attracted relatively little attention. In this paper, we describe a hardware-efficient on-line learning technique for feedforward multi-layer ANNs that is based on pipelined backpropagation. Learning is performed in parallel with inference in the forward pass, removing the need for an explicit backward pass and requiring no extra weight lookup. By using binary state variables in the feedforward network and ternary errors in truncated-error backpropagation, the need for any multiplications in the forward and backward passes is removed, and memory requirements for the pipelining are drastically reduced. Further reduction in addition operations owing to the sparsity in the forward neural and backpropagating error signal paths contributes to highly efficient hardware implementation. For proof-of-concept validation, we demonstrate on-line learning of MNIST handwritten digit classification on a Spartan 6 FPGA interfacing with an external 1Gb DDR2 DRAM, that shows small degradation in test error performance compared to an equivalently sized binary ANN trained off-line using standard back-propagation and exact errors. Our results highlight an attractive synergy between pipelined backpropagation and binary-state networks in substantially reducing computation and memory requirements, making pipelined on-line learning practical in deep networks.
Reliability of a treatment-based classification system for subgrouping people with low back pain.
Henry, Sharon M; Fritz, Julie M; Trombley, Andrea R; Bunn, Janice Y
2012-09-01
Observational, cross-sectional reliability study. To examine the interrater reliability of novice raters in their use of the treatment-based classification (TBC) system for low back pain and to explore the patterns of disagreement in classification errors. Although the interrater reliability of individual test items in the TBC system is moderate to good, some error persists in classification decision making. Understanding which classification errors are common could direct further refinement of the TBC system. Using previously recorded patient data (n = 24), 12 novice raters classified patients according to the TBC schema. These classification results were combined with those of 7 other raters, allowing examination of the overall agreement using the kappa statistic, as well as agreement/disagreement among pairwise comparisons in classification assignments. A chi-square test examined differences in percent agreement between the novice and more experienced raters and differences in classification distributions between these 2 groups of raters. Among 12 novice raters, there was 80.9% agreement in the pairs of classification (κ = 0.62; 95% confidence interval: 0.59, 0.65) and an overall 75.5% agreement (κ = 0.57; 95% confidence interval: 0.55, 0.69) for the combined data set. Raters were least likely to agree on a classification of stabilization (77.5% agreement). The overall percentage of pairwise classification judgments that disagreed was 24.5%, with the most common disagreement being between manipulation and stabilization (11.0%), followed by a mismatch between stabilization and specific exercise (8.2%). Additional refinement is needed to reduce rater disagreement that persists in the TBC decision-making algorithm, particularly in the stabilization category. J Orthop Sports Phys Ther 2012;42(9):797-805, Epub 7 June 2012. doi:10.2519/jospt.2012.4078.
Hot complaint intelligent classification based on text mining
Directory of Open Access Journals (Sweden)
XIA Haifeng
2013-10-01
Full Text Available The complaint recognizer system plays an important role in making sure the correct classification of the hot complaint,improving the service quantity of telecommunications industry.The customers’ complaint in telecommunications industry has its special particularity which should be done in limited time,which cause the error in classification of hot complaint.The paper presents a model of complaint hot intelligent classification based on text mining,which can classify the hot complaint in the correct level of the complaint navigation.The examples show that the model can be efficient to classify the text of the complaint.
Nursing Errors in Intensive Care Unit by Human Error Identification in Systems Tool: A Case Study
Directory of Open Access Journals (Sweden)
Nezamodini
2016-03-01
Full Text Available Background Although health services are designed and implemented to improve human health, the errors in health services are a very common phenomenon and even sometimes fatal in this field. Medical errors and their cost are global issues with serious consequences for the patients’ community that are preventable and require serious attention. Objectives The current study aimed to identify possible nursing errors applying human error identification in systems tool (HEIST in the intensive care units (ICUs of hospitals. Patients and Methods This descriptive research was conducted in the intensive care unit of a hospital in Khuzestan province in 2013. Data were collected through observation and interview by nine nurses in this section in a period of four months. Human error classification was based on Rose and Rose and Swain and Guttmann models. According to HEIST work sheets the guide questions were answered and error causes were identified after the determination of the type of errors. Results In total 527 errors were detected. The performing operation on the wrong path had the highest frequency which was 150, and the second rate with a frequency of 136 was doing the tasks later than the deadline. Management causes with a frequency of 451 were the first rank among identified errors. Errors mostly occurred in the system observation stage and among the performance shaping factors (PSFs, time was the most influencing factor in occurrence of human errors. Conclusions Finally, in order to prevent the occurrence and reduce the consequences of identified errors the following suggestions were proposed : appropriate training courses, applying work guidelines and monitoring their implementation, increasing the number of work shifts, hiring professional workforce, equipping work space with appropriate facilities and equipment.
Blind Signal Classification via Spare Coding
2016-04-10
Blind Signal Classification via Sparse Coding Youngjune Gwon MIT Lincoln Laboratory gyj@ll.mit.edu Siamak Dastangoo MIT Lincoln Laboratory sia...achieve blind signal classification with no prior knowledge about signals (e.g., MCS, pulse shaping) in an arbitrary RF channel. Since modulated RF...classification method. Our results indicate that we can separate different classes of digitally modulated signals from blind sampling with 70.3% recall and 24.6
An Improved Rotation Forest for Multi-Feature Remote-Sensing Imagery Classification
Directory of Open Access Journals (Sweden)
Yingchang Xiu
2017-11-01
Full Text Available Multi-feature, especially multi-temporal, remote-sensing data have the potential to improve land cover classification accuracy. However, sometimes it is difficult to utilize all the features efficiently. To enhance classification performance based on multi-feature imagery, an improved rotation forest, combining Principal Component Analysis (PCA and a boosting naïve Bayesian tree (NBTree, is proposed. First, feature extraction was carried out with PCA. The feature set was randomly split into several disjoint subsets; then, PCA was applied to each subset, and new training data for linear extracted features based on original training data were obtained. These steps were repeated several times. Second, based on the new training data, a boosting naïve Bayesian tree was constructed as the base classifier, which aims to achieve lower prediction error than a decision tree in the original rotation forest. At the classification phase, the improved rotation forest has two-layer voting. It first obtains several predictions through weighted voting in a boosting naïve Bayesian tree; then, the first-layer vote predicts by majority to obtain the final result. To examine the classification performance, the improved rotation forest was applied to multi-feature remote-sensing images, including MODIS Enhanced Vegetation Index (EVI imagery time series, MODIS Surface Reflectance products and ancillary data in Shandong Province for 2013. The EVI imagery time series was preprocessed using harmonic analysis of time series (HANTS to reduce the noise effects. The overall accuracy of the final classification result was 89.17%, and the Kappa coefficient was 0.71, which outperforms the original rotation forest and other classifier ensemble results, as well as the NASA land cover product. However, this new algorithm requires more computational time, meaning the efficiency needs to be further improved. Generally, the improved rotation forest has a potential advantage in
Fusing metabolomics data sets with heterogeneous measurement errors
Waaijenborg, Sandra; Korobko, Oksana; Willems van Dijk, Ko; Lips, Mirjam; Hankemeier, Thomas; Wilderjans, Tom F.; Smilde, Age K.
2018-01-01
Combining different metabolomics platforms can contribute significantly to the discovery of complementary processes expressed under different conditions. However, analysing the fused data might be hampered by the difference in their quality. In metabolomics data, one often observes that measurement errors increase with increasing measurement level and that different platforms have different measurement error variance. In this paper we compare three different approaches to correct for the measurement error heterogeneity, by transformation of the raw data, by weighted filtering before modelling and by a modelling approach using a weighted sum of residuals. For an illustration of these different approaches we analyse data from healthy obese and diabetic obese individuals, obtained from two metabolomics platforms. Concluding, the filtering and modelling approaches that both estimate a model of the measurement error did not outperform the data transformation approaches for this application. This is probably due to the limited difference in measurement error and the fact that estimation of measurement error models is unstable due to the small number of repeats available. A transformation of the data improves the classification of the two groups. PMID:29698490
Three-Class Mammogram Classification Based on Descriptive CNN Features
Directory of Open Access Journals (Sweden)
M. Mohsin Jadoon
2017-01-01
Full Text Available In this paper, a novel classification technique for large data set of mammograms using a deep learning method is proposed. The proposed model targets a three-class classification study (normal, malignant, and benign cases. In our model we have presented two methods, namely, convolutional neural network-discrete wavelet (CNN-DW and convolutional neural network-curvelet transform (CNN-CT. An augmented data set is generated by using mammogram patches. To enhance the contrast of mammogram images, the data set is filtered by contrast limited adaptive histogram equalization (CLAHE. In the CNN-DW method, enhanced mammogram images are decomposed as its four subbands by means of two-dimensional discrete wavelet transform (2D-DWT, while in the second method discrete curvelet transform (DCT is used. In both methods, dense scale invariant feature (DSIFT for all subbands is extracted. Input data matrix containing these subband features of all the mammogram patches is created that is processed as input to convolutional neural network (CNN. Softmax layer and support vector machine (SVM layer are used to train CNN for classification. Proposed methods have been compared with existing methods in terms of accuracy rate, error rate, and various validation assessment measures. CNN-DW and CNN-CT have achieved accuracy rate of 81.83% and 83.74%, respectively. Simulation results clearly validate the significance and impact of our proposed model as compared to other well-known existing techniques.
Kreilinger, Alex; Hiebel, Hannah; Müller-Putz, Gernot R
2016-03-01
This work aimed to find and evaluate a new method for detecting errors in continuous brain-computer interface (BCI) applications. Instead of classifying errors on a single-trial basis, the new method was based on multiple events (MEs) analysis to increase the accuracy of error detection. In a BCI-driven car game, based on motor imagery (MI), discrete events were triggered whenever subjects collided with coins and/or barriers. Coins counted as correct events, whereas barriers were errors. This new method, termed ME method, combined and averaged the classification results of single events (SEs) and determined the correctness of MI trials, which consisted of event sequences instead of SEs. The benefit of this method was evaluated in an offline simulation. In an online experiment, the new method was used to detect erroneous MI trials. Such MI trials were discarded and could be repeated by the users. We found that, even with low SE error potential (ErrP) detection rates, feasible accuracies can be achieved when combining MEs to distinguish erroneous from correct MI trials. Online, all subjects reached higher scores with error detection than without, at the cost of longer times needed for completing the game. Findings suggest that ErrP detection may become a reliable tool for monitoring continuous states in BCI applications when combining MEs. This paper demonstrates a novel technique for detecting errors in online continuous BCI applications, which yields promising results even with low single-trial detection rates.
Spectral classification of emission-line galaxies
International Nuclear Information System (INIS)
Veilleux, S.; Osterbrock, D.E.
1987-01-01
A revised method of classification of narrow-line active galaxies and H II region-like galaxies is proposed. It involves the line ratios which take full advantage of the physical distinction between the two types of objects and minimize the effects of reddening correction and errors in the flux calibration. Large sets of internally consistent data are used, including new, previously unpublished measurements. Predictions of recent photoionization models by power-law spectra and by hot stars are compared with the observations. The classification is based on the observational data interpreted on the basis of these models. 63 references
Using Errors to Improve the Quality of Instructional Programs.
Anderson, Lorin W.; And Others
Clinchy and Rosenthal's error classification scheme was applied to test results to determine its ability to differentiate the effectiveness of instruction in two elementary schools. Mathematics retention tests matching the instructional objectives of both schools were constructed to measure the understanding of arithmetic concepts and the ability…
Photon level chemical classification using digital compressive detection
International Nuclear Information System (INIS)
Wilcox, David S.; Buzzard, Gregery T.; Lucier, Bradley J.; Wang Ping; Ben-Amotz, Dor
2012-01-01
Highlights: ► A new digital compressive detection strategy is developed. ► Chemical classification demonstrated using as few as ∼10 photons. ► Binary filters are optimal when taking few measurements. - Abstract: A key bottleneck to high-speed chemical analysis, including hyperspectral imaging and monitoring of dynamic chemical processes, is the time required to collect and analyze hyperspectral data. Here we describe, both theoretically and experimentally, a means of greatly speeding up the collection of such data using a new digital compressive detection strategy. Our results demonstrate that detecting as few as ∼10 Raman scattered photons (in as little time as ∼30 μs) can be sufficient to positively distinguish chemical species. This is achieved by measuring the Raman scattered light intensity transmitted through programmable binary optical filters designed to minimize the error in the chemical classification (or concentration) variables of interest. The theoretical results are implemented and validated using a digital compressive detection instrument that incorporates a 785 nm diode excitation laser, digital micromirror spatial light modulator, and photon counting photodiode detector. Samples consisting of pairs of liquids with different degrees of spectral overlap (including benzene/acetone and n-heptane/n-octane) are used to illustrate how the accuracy of the present digital compressive detection method depends on the correlation coefficients of the corresponding spectra. Comparisons of measured and predicted chemical classification score plots, as well as linear and non-linear discriminant analyses, demonstrate that this digital compressive detection strategy is Poisson photon noise limited and outperforms total least squares-based compressive detection with analog filters.
Lewis, Katherine E.
2016-01-01
Although many students struggle with fractions, students with mathematical learning disabilities (MLDs) experience pervasive difficulties because of neurological differences in how they process numerical information. These students make errors that are qualitatively different than their typically achieving and low-achieving peers. This study…
Effects of stress typicality during speeded grammatical classification.
Arciuli, Joanne; Cupples, Linda
2003-01-01
The experiments reported here were designed to investigate the influence of stress typicality during speeded grammatical classification of disyllabic English words by native and non-native speakers. Trochaic nouns and iambic gram verbs were considered to be typically stressed, whereas iambic nouns and trochaic verbs were considered to be atypically stressed. Experiments 1a and 2a showed that while native speakers classified typically stressed words individual more quickly and more accurately than atypically stressed words during differences reading, there were no overall effects during classification of spoken stimuli. However, a subgroup of native speakers with high error rates did show a significant effect during classification of spoken stimuli. Experiments 1b and 2b showed that non-native speakers classified typically stressed words more quickly and more accurately than atypically stressed words during reading. Typically stressed words were classified more accurately than atypically stressed words when the stimuli were spoken. Importantly, there was a significant relationship between error rates, vocabulary size and the size of the stress typicality effect in each experiment. We conclude that participants use information about lexical stress to help them distinguish between disyllabic nouns and verbs during speeded grammatical classification. This is especially so for individuals with a limited vocabulary who lack other knowledge (e.g., semantic knowledge) about the differences between these grammatical categories.
Co-occurrence Models in Music Genre Classification
DEFF Research Database (Denmark)
Ahrendt, Peter; Goutte, Cyril; Larsen, Jan
2005-01-01
Music genre classification has been investigated using many different methods, but most of them build on probabilistic models of feature vectors x\\_r which only represent the short time segment with index r of the song. Here, three different co-occurrence models are proposed which instead consider...... genre data set with a variety of modern music. The basis was a so-called AR feature representation of the music. Besides the benefit of having proper probabilistic models of the whole song, the lowest classification test errors were found using one of the proposed models....
Automatic classification of background EEG activity in healthy and sick neonates
Löfhede, Johan; Thordstein, Magnus; Löfgren, Nils; Flisberg, Anders; Rosa-Zurera, Manuel; Kjellmer, Ingemar; Lindecrantz, Kaj
2010-02-01
The overall aim of our research is to develop methods for a monitoring system to be used at neonatal intensive care units. When monitoring a baby, a range of different types of background activity needs to be considered. In this work, we have developed a scheme for automatic classification of background EEG activity in newborn babies. EEG from six full-term babies who were displaying a burst suppression pattern while suffering from the after-effects of asphyxia during birth was included along with EEG from 20 full-term healthy newborn babies. The signals from the healthy babies were divided into four behavioural states: active awake, quiet awake, active sleep and quiet sleep. By using a number of features extracted from the EEG together with Fisher's linear discriminant classifier we have managed to achieve 100% correct classification when separating burst suppression EEG from all four healthy EEG types and 93% true positive classification when separating quiet sleep from the other types. The other three sleep stages could not be classified. When the pathological burst suppression pattern was detected, the analysis was taken one step further and the signal was segmented into burst and suppression, allowing clinically relevant parameters such as suppression length and burst suppression ratio to be calculated. The segmentation of the burst suppression EEG works well, with a probability of error around 4%.
The generalization ability of online SVM classification based on Markov sampling.
Xu, Jie; Yan Tang, Yuan; Zou, Bin; Xu, Zongben; Li, Luoqing; Lu, Yang
2015-03-01
In this paper, we consider online support vector machine (SVM) classification learning algorithms with uniformly ergodic Markov chain (u.e.M.c.) samples. We establish the bound on the misclassification error of an online SVM classification algorithm with u.e.M.c. samples based on reproducing kernel Hilbert spaces and obtain a satisfactory convergence rate. We also introduce a novel online SVM classification algorithm based on Markov sampling, and present the numerical studies on the learning ability of online SVM classification based on Markov sampling for benchmark repository. The numerical studies show that the learning performance of the online SVM classification algorithm based on Markov sampling is better than that of classical online SVM classification based on random sampling as the size of training samples is larger.
Assessing Measures of Order Flow Toxicity via Perfect Trade Classification
DEFF Research Database (Denmark)
Andersen, Torben G.; Bondarenko, Oleg
. The VPIN metric involves decomposing volume into active buys and sells. We use the best-bid-offer (BBO) files from the CME Group to construct (near) perfect trade classification measures for the E-mini S&P 500 futures contract. We investigate the accuracy of the ELO Bulk Volume Classification (BVC) scheme...... systematic classification errors that are correlated with trading volume and return volatility. When controlling for trading intensity and volatility, the BVC-VPIN measure has no incremental predictive power for future volatility. We conclude that VPIN is not suitable for measuring order flow imbalances....
Managing Sensitive Information: DOD Can More Effectively Reduce the Risk of Classification Errors
National Research Council Canada - National Science Library
D'Agostino, Davi M; Borseth, Ann; Fenton, Mattias; Hatton, Adam; Hills, Barbara; Keefer, David; Mayfield, David; Reid, Jim; Richardson, Terry; Schwartz, Marc
2006-01-01
.... While some DoD components and their subordinate commands appear to manage effective programs, GAO identified weaknesses in others in the areas of classification management training, self-inspections...
The error model and experiment of measuring angular position error based on laser collimation
Cai, Yangyang; Yang, Jing; Li, Jiakun; Feng, Qibo
2018-01-01
Rotary axis is the reference component of rotation motion. Angular position error is the most critical factor which impair the machining precision among the six degree-of-freedom (DOF) geometric errors of rotary axis. In this paper, the measuring method of angular position error of rotary axis based on laser collimation is thoroughly researched, the error model is established and 360 ° full range measurement is realized by using the high precision servo turntable. The change of space attitude of each moving part is described accurately by the 3×3 transformation matrices and the influences of various factors on the measurement results is analyzed in detail. Experiments results show that the measurement method can achieve high measurement accuracy and large measurement range.
Cough event classification by pretrained deep neural network.
Liu, Jia-Ming; You, Mingyu; Wang, Zheng; Li, Guo-Zheng; Xu, Xianghuai; Qiu, Zhongmin
2015-01-01
Cough is an essential symptom in respiratory diseases. In the measurement of cough severity, an accurate and objective cough monitor is expected by respiratory disease society. This paper aims to introduce a better performed algorithm, pretrained deep neural network (DNN), to the cough classification problem, which is a key step in the cough monitor. The deep neural network models are built from two steps, pretrain and fine-tuning, followed by a Hidden Markov Model (HMM) decoder to capture tamporal information of the audio signals. By unsupervised pretraining a deep belief network, a good initialization for a deep neural network is learned. Then the fine-tuning step is a back propogation tuning the neural network so that it can predict the observation probability associated with each HMM states, where the HMM states are originally achieved by force-alignment with a Gaussian Mixture Model Hidden Markov Model (GMM-HMM) on the training samples. Three cough HMMs and one noncough HMM are employed to model coughs and noncoughs respectively. The final decision is made based on viterbi decoding algorihtm that generates the most likely HMM sequence for each sample. A sample is labeled as cough if a cough HMM is found in the sequence. The experiments were conducted on a dataset that was collected from 22 patients with respiratory diseases. Patient dependent (PD) and patient independent (PI) experimental settings were used to evaluate the models. Five criteria, sensitivity, specificity, F1, macro average and micro average are shown to depict different aspects of the models. From overall evaluation criteria, the DNN based methods are superior to traditional GMM-HMM based method on F1 and micro average with maximal 14% and 11% error reduction in PD and 7% and 10% in PI, meanwhile keep similar performances on macro average. They also surpass GMM-HMM model on specificity with maximal 14% error reduction on both PD and PI. In this paper, we tried pretrained deep neural network in
AN ANALYSIS OF ACEHNESE EFL STUDENTS’ GRAMMATICAL ERRORS IN WRITING RECOUNT TEXTS
Directory of Open Access Journals (Sweden)
Qudwatin Nisak M. Isa
2017-11-01
Full Text Available This study aims at finding empirical evidence of the most common types of grammatical errors and sources of errors in recount texts written by the first-year students of SMAS Babul Maghfirah, Aceh Besar. The subject of the study was a collection of students’ personal writing documents of recount texts about their lives experience. The students’ recount texts were analyzed by referring to Betty S. Azar classification and Richard’s theory on sources of errors. The findings showed that the total number of error is 436. Two frequent types of grammatical errors were Verb Tense and Word Choice. The major sources of error were Intralingual Error, Interference Error and Developmental Error respectively. Furthermore, the findings suggest that it is necessary for EFL teachers to apply appropriate techniques and strategies in teaching recount texts, which focus on past tense and language features of the text in order to reduce the possible errors to be made by the students.
Recognition of medical errors' reporting system dimensions in educational hospitals.
Yarmohammadian, Mohammad H; Mohammadinia, Leila; Tavakoli, Nahid; Ghalriz, Parvin; Haghshenas, Abbas
2014-01-01
Nowadays medical errors are one of the serious issues in the health-care system and carry to account of the patient's safety threat. The most important step for achieving safety promotion is identifying errors and their causes in order to recognize, correct and omit them. Concerning about repeating medical errors and harms, which were received via theses errors concluded to designing and establishing medical error reporting systems for hospitals and centers that are presenting therapeutic services. The aim of this study is the recognition of medical errors' reporting system dimensions in educational hospitals. This research is a descriptive-analytical and qualities' study, which has been carried out in Shahid Beheshti educational therapeutic center in Isfahan during 2012. In this study, relevant information was collected through 15 face to face interviews. That each of interviews take place in about 1hr and creation of five focused discussion groups through 45 min for each section, they were composed of Metron, educational supervisor, health officer, health education, and all of the head nurses. Concluded data interviews and discussion sessions were coded, then achieved results were extracted in the presence of clear-sighted persons and after their feedback perception, they were categorized. In order to make sure of information correctness, tables were presented to the research's interviewers and final the corrections were confirmed based on their view. The extracted information from interviews and discussion groups have been divided into nine main categories after content analyzing and subject coding and their subsets have been completely expressed. Achieved dimensions are composed of nine domains of medical error concept, error cases according to nurses' prospection, medical error reporting barriers, employees' motivational factors for error reporting, purposes of medical error reporting system, error reporting's challenges and opportunities, a desired system
International Nuclear Information System (INIS)
Richards, Joseph W.; Starr, Dan L.; Miller, Adam A.; Bloom, Joshua S.; Brink, Henrik; Crellin-Quick, Arien; Butler, Nathaniel R.
2012-01-01
With growing data volumes from synoptic surveys, astronomers necessarily must become more abstracted from the discovery and introspection processes. Given the scarcity of follow-up resources, there is a particularly sharp onus on the frameworks that replace these human roles to provide accurate and well-calibrated probabilistic classification catalogs. Such catalogs inform the subsequent follow-up, allowing consumers to optimize the selection of specific sources for further study and permitting rigorous treatment of classification purities and efficiencies for population studies. Here, we describe a process to produce a probabilistic classification catalog of variability with machine learning from a multi-epoch photometric survey. In addition to producing accurate classifications, we show how to estimate calibrated class probabilities and motivate the importance of probability calibration. We also introduce a methodology for feature-based anomaly detection, which allows discovery of objects in the survey that do not fit within the predefined class taxonomy. Finally, we apply these methods to sources observed by the All-Sky Automated Survey (ASAS), and release the Machine-learned ASAS Classification Catalog (MACC), a 28 class probabilistic classification catalog of 50,124 ASAS sources in the ASAS Catalog of Variable Stars. We estimate that MACC achieves a sub-20% classification error rate and demonstrate that the class posterior probabilities are reasonably calibrated. MACC classifications compare favorably to the classifications of several previous domain-specific ASAS papers and to the ASAS Catalog of Variable Stars, which had classified only 24% of those sources into one of 12 science classes.
Energy Technology Data Exchange (ETDEWEB)
Richards, Joseph W.; Starr, Dan L.; Miller, Adam A.; Bloom, Joshua S.; Brink, Henrik; Crellin-Quick, Arien [Astronomy Department, University of California, Berkeley, CA 94720-3411 (United States); Butler, Nathaniel R., E-mail: jwrichar@stat.berkeley.edu [School of Earth and Space Exploration, Arizona State University, Tempe, AZ 85287 (United States)
2012-12-15
With growing data volumes from synoptic surveys, astronomers necessarily must become more abstracted from the discovery and introspection processes. Given the scarcity of follow-up resources, there is a particularly sharp onus on the frameworks that replace these human roles to provide accurate and well-calibrated probabilistic classification catalogs. Such catalogs inform the subsequent follow-up, allowing consumers to optimize the selection of specific sources for further study and permitting rigorous treatment of classification purities and efficiencies for population studies. Here, we describe a process to produce a probabilistic classification catalog of variability with machine learning from a multi-epoch photometric survey. In addition to producing accurate classifications, we show how to estimate calibrated class probabilities and motivate the importance of probability calibration. We also introduce a methodology for feature-based anomaly detection, which allows discovery of objects in the survey that do not fit within the predefined class taxonomy. Finally, we apply these methods to sources observed by the All-Sky Automated Survey (ASAS), and release the Machine-learned ASAS Classification Catalog (MACC), a 28 class probabilistic classification catalog of 50,124 ASAS sources in the ASAS Catalog of Variable Stars. We estimate that MACC achieves a sub-20% classification error rate and demonstrate that the class posterior probabilities are reasonably calibrated. MACC classifications compare favorably to the classifications of several previous domain-specific ASAS papers and to the ASAS Catalog of Variable Stars, which had classified only 24% of those sources into one of 12 science classes.
Optimizing tree-species classification in hyperspectal images
CSIR Research Space (South Africa)
Barnard, E
2010-11-01
Full Text Available for classification. Scaling of these components so that all features have equal variance is found to be useful, and their best performance (88.9% accurate classification) is achieved with 15 scaled features and a support vector machine as classifier. A graphical...
Credit Risk Evaluation Using a C-Variable Least Squares Support Vector Classification Model
Yu, Lean; Wang, Shouyang; Lai, K. K.
Credit risk evaluation is one of the most important issues in financial risk management. In this paper, a C-variable least squares support vector classification (C-VLSSVC) model is proposed for credit risk analysis. The main idea of this model is based on the prior knowledge that different classes may have different importance for modeling and more weights should be given to those classes with more importance. The C-VLSSVC model can be constructed by a simple modification of the regularization parameter in LSSVC, whereby more weights are given to the lease squares classification errors with important classes than the lease squares classification errors with unimportant classes while keeping the regularized terms in its original form. For illustration purpose, a real-world credit dataset is used to test the effectiveness of the C-VLSSVC model.
Pietrzyk, Mariusz W.; Donovan, Tim; Brennan, Patrick C.; Dix, Alan; Manning, David J.
2011-03-01
Aim: To optimize automated classification of radiological errors during lung nodule detection from chest radiographs (CxR) using a support vector machine (SVM) run on the spatial frequency features extracted from the local background of selected regions. Background: The majority of the unreported pulmonary nodules are visually detected but not recognized; shown by the prolonged dwell time values at false-negative regions. Similarly, overestimated nodule locations are capturing substantial amounts of foveal attention. Spatial frequency properties of selected local backgrounds are correlated with human observer responses either in terms of accuracy in indicating abnormality position or in the precision of visual sampling the medical images. Methods: Seven radiologists participated in the eye tracking experiments conducted under conditions of pulmonary nodule detection from a set of 20 postero-anterior CxR. The most dwelled locations have been identified and subjected to spatial frequency (SF) analysis. The image-based features of selected ROI were extracted with un-decimated Wavelet Packet Transform. An analysis of variance was run to select SF features and a SVM schema was implemented to classify False-Negative and False-Positive from all ROI. Results: A relative high overall accuracy was obtained for each individually developed Wavelet-SVM algorithm, with over 90% average correct ratio for errors recognition from all prolonged dwell locations. Conclusion: The preliminary results show that combined eye-tracking and image-based features can be used for automated detection of radiological error with SVM. The work is still in progress and not all analytical procedures have been completed, which might have an effect on the specificity of the algorithm.
International Nuclear Information System (INIS)
Nidaira, Kazuo
2008-01-01
International Target Values (ITV) shows random and systematic measurement uncertainty components as a reference for routinely achievable measurement quality in the accountancy measurement. The measurement uncertainty, called error henceforth, needs to be periodically evaluated and checked against ITV for consistency as the error varies according to measurement methods, instruments, operators, certified reference samples, frequency of calibration, and so on. In the paper an error evaluation method was developed with focuses on (1) Specifying clearly error calculation model, (2) Getting always positive random and systematic error variances, (3) Obtaining probability density distribution of an error variance and (4) Confirming the evaluation method by simulation. In addition the method was demonstrated by applying real data. (author)
Classification system for reporting events involving human malfunctions
DEFF Research Database (Denmark)
Rasmussen, Jens; Pedersen, O.M.; Mancini, G.
1981-01-01
The report describes a set of categories for reporting indus-trial incidents and events involving human malfunction. The classification system aims at ensuring information adequate for improvement of human work situations and man-machine interface systems and for attempts to quantify "human error......" rates. The classification system has a multifacetted non-hierarchical struc-ture and its compatibility with Isprals ERDS classification is described. The collection of the information in general and for quantification purposes are discussed. 24 categories, 12 of which being human factors oriented......, are listed with their respective subcategories, and comments are given. Underlying models of human data processes and their typical malfunc-tions and of a human decision sequence are described....
Classification system for reporting events involving human malfunctions
International Nuclear Information System (INIS)
Rasmussen, J.; Pedersen, O.M.; Mancini, G.; Carnino, A.; Griffon, M.; Gagnolet, P.
1981-03-01
The report describes a set of categories for reporting industrial incidents and events involving human malfunction. The classification system aims at ensuring information adequate for improvement of human work situations and man-machine interface systems and for attempts to quantify ''human error'' rates. The classification system has a multifacetted non-hierarchial structure and its compatibility with Ispra's ERDS classification is described. The collection of the information in general and for quantification purposes are discussed. 24 categories, 12 of which being human factors oriented, are listed with their respective subcategories, and comments are given. Underlying models of human data processes and their typical malfunctions and of a human decision sequence are described. (author)
Introduction to precision machine design and error assessment
Mekid, Samir
2008-01-01
While ultra-precision machines are now achieving sub-nanometer accuracy, unique challenges continue to arise due to their tight specifications. Written to meet the growing needs of mechanical engineers and other professionals to understand these specialized design process issues, Introduction to Precision Machine Design and Error Assessment places a particular focus on the errors associated with precision design, machine diagnostics, error modeling, and error compensation. Error Assessment and ControlThe book begins with a brief overview of precision engineering and applications before introdu
Carter, Jeffrey R.; Simon, Wayne E.
1990-08-01
Neural networks are trained using Recursive Error Minimization (REM) equations to perform statistical classification. Using REM equations with continuous input variables reduces the required number of training experiences by factors of one to two orders of magnitude over standard back propagation. Replacing the continuous input variables with discrete binary representations reduces the number of connections by a factor proportional to the number of variables reducing the required number of experiences by another order of magnitude. Undesirable effects of using recurrent experience to train neural networks for statistical classification problems are demonstrated and nonrecurrent experience used to avoid these undesirable effects. 1. THE 1-41 PROBLEM The statistical classification problem which we address is is that of assigning points in ddimensional space to one of two classes. The first class has a covariance matrix of I (the identity matrix) the covariance matrix of the second class is 41. For this reason the problem is known as the 1-41 problem. Both classes have equal probability of occurrence and samples from both classes may appear anywhere throughout the ddimensional space. Most samples near the origin of the coordinate system will be from the first class while most samples away from the origin will be from the second class. Since the two classes completely overlap it is impossible to have a classifier with zero error. The minimum possible error is known as the Bayes error and
Directory of Open Access Journals (Sweden)
Gabriella Rejane dos Santos Dalmolin
2013-12-01
Full Text Available Medication errors can be frequent in hospitals; these errors are multidisciplinary and occur at various stages of the drug therapy. The present study evaluated the seriousness, the type and the drugs involved in medication errors reported at the Hospital de Clínicas de Porto Alegre. We analyzed written error reports for 2010-2011. The sample consisted of 165 reports. The errors identified were classified according to seriousness, type and pharmacological class. 114 reports were categorized as actual errors (medication errors and 51 reports were categorized as potential errors. There were more medication error reports in 2011 compared to 2010, but there was no significant change in the seriousness of the reports. The most common type of error was prescribing error (48.25%. Errors that occurred during the process of drug therapy sometimes generated additional medication errors. In 114 reports of medication errors identified, 122 drugs were cited. The reflection on medication errors, the possibility of harm resulting from these errors, and the methods for error identification and evaluation should include a broad perspective of the aspects involved in the occurrence of errors. Patient safety depends on the process of communication involving errors, on the proper recording of information, and on the monitoring itself.
Directory of Open Access Journals (Sweden)
Carolina Reta
Full Text Available Morphological identification of acute leukemia is a powerful tool used by hematologists to determine the family of such a disease. In some cases, experienced physicians are even able to determine the leukemia subtype of the sample. However, the identification process may have error rates up to 40% (when classifying acute leukemia subtypes depending on the physician's experience and the sample quality. This problem raises the need to create automatic tools that provide hematologists with a second opinion during the classification process. Our research presents a contextual analysis methodology for the detection of acute leukemia subtypes from bone marrow cells images. We propose a cells separation algorithm to break up overlapped regions. In this phase, we achieved an average accuracy of 95% in the evaluation of the segmentation process. In a second phase, we extract descriptive features to the nucleus and cytoplasm obtained in the segmentation phase in order to classify leukemia families and subtypes. We finally created a decision algorithm that provides an automatic diagnosis for a patient. In our experiments, we achieved an overall accuracy of 92% in the supervised classification of acute leukemia families, 84% for the lymphoblastic subtypes, and 92% for the myeloblastic subtypes. Finally, we achieved accuracies of 95% in the diagnosis of leukemia families and 90% in the diagnosis of leukemia subtypes.
High Dimensional Classification Using Features Annealed Independence Rules.
Fan, Jianqing; Fan, Yingying
2008-01-01
Classification using high-dimensional features arises frequently in many contemporary statistical studies such as tumor classification using microarray or other high-throughput data. The impact of dimensionality on classifications is largely poorly understood. In a seminal paper, Bickel and Levina (2004) show that the Fisher discriminant performs poorly due to diverging spectra and they propose to use the independence rule to overcome the problem. We first demonstrate that even for the independence classification rule, classification using all the features can be as bad as the random guessing due to noise accumulation in estimating population centroids in high-dimensional feature space. In fact, we demonstrate further that almost all linear discriminants can perform as bad as the random guessing. Thus, it is paramountly important to select a subset of important features for high-dimensional classification, resulting in Features Annealed Independence Rules (FAIR). The conditions under which all the important features can be selected by the two-sample t-statistic are established. The choice of the optimal number of features, or equivalently, the threshold value of the test statistics are proposed based on an upper bound of the classification error. Simulation studies and real data analysis support our theoretical results and demonstrate convincingly the advantage of our new classification procedure.
International Nuclear Information System (INIS)
Hirotsu, Yuko; Tsuge, Tadashi; Sano, Toshiaki; Takano, Kenichi; Gouda, Hidenori
2001-01-01
CRIEPI has been conducting detailed analyses of all human error incidents at domestic nuclear power plants (NPPs) collected from Japanese Licensee Event Reports (LERs) using J-HPES (Japanese version of HPES) as an analysis method. Results obtained by the analyses have been stored in J-HPES database. Since 1999, human error incidents have been selected from U.S. LERs, and they are analyzed using J-HPES. In this report, the results, which classified error action, cause, and preventive measure, are summarized for U.S. human error cases occurring in the period from 1992 to 1996. It was suggested as a result of classification that the categories of error action were almost the same as those of Japanese human error cases. Therefore, problems in the process of error action and checkpoints for preventing errors will be extracted by analyzing both U.S. and domestic human error cases. It was also suggested that the interrelations between error actions, causes, and organizational factors could be identified. While taking these suggestions into consideration, we will continue to analyze U.S. human error cases. (author)
Influence of nuclei segmentation on breast cancer malignancy classification
Jelen, Lukasz; Fevens, Thomas; Krzyzak, Adam
2009-02-01
Breast Cancer is one of the most deadly cancers affecting middle-aged women. Accurate diagnosis and prognosis are crucial to reduce the high death rate. Nowadays there are numerous diagnostic tools for breast cancer diagnosis. In this paper we discuss a role of nuclear segmentation from fine needle aspiration biopsy (FNA) slides and its influence on malignancy classification. Classification of malignancy plays a very important role during the diagnosis process of breast cancer. Out of all cancer diagnostic tools, FNA slides provide the most valuable information about the cancer malignancy grade which helps to choose an appropriate treatment. This process involves assessing numerous nuclear features and therefore precise segmentation of nuclei is very important. In this work we compare three powerful segmentation approaches and test their impact on the classification of breast cancer malignancy. The studied approaches involve level set segmentation, fuzzy c-means segmentation and textural segmentation based on co-occurrence matrix. Segmented nuclei were used to extract nuclear features for malignancy classification. For classification purposes four different classifiers were trained and tested with previously extracted features. The compared classifiers are Multilayer Perceptron (MLP), Self-Organizing Maps (SOM), Principal Component-based Neural Network (PCA) and Support Vector Machines (SVM). The presented results show that level set segmentation yields the best results over the three compared approaches and leads to a good feature extraction with a lowest average error rate of 6.51% over four different classifiers. The best performance was recorded for multilayer perceptron with an error rate of 3.07% using fuzzy c-means segmentation.
Compensatory neurofuzzy model for discrete data classification in biomedical
Ceylan, Rahime
2015-03-01
Biomedical data is separated to two main sections: signals and discrete data. So, studies in this area are about biomedical signal classification or biomedical discrete data classification. There are artificial intelligence models which are relevant to classification of ECG, EMG or EEG signals. In same way, in literature, many models exist for classification of discrete data taken as value of samples which can be results of blood analysis or biopsy in medical process. Each algorithm could not achieve high accuracy rate on classification of signal and discrete data. In this study, compensatory neurofuzzy network model is presented for classification of discrete data in biomedical pattern recognition area. The compensatory neurofuzzy network has a hybrid and binary classifier. In this system, the parameters of fuzzy systems are updated by backpropagation algorithm. The realized classifier model is conducted to two benchmark datasets (Wisconsin Breast Cancer dataset and Pima Indian Diabetes dataset). Experimental studies show that compensatory neurofuzzy network model achieved 96.11% accuracy rate in classification of breast cancer dataset and 69.08% accuracy rate was obtained in experiments made on diabetes dataset with only 10 iterations.
Practical, Reliable Error Bars in Quantum Tomography
Faist, Philippe; Renner, Renato
2015-01-01
Precise characterization of quantum devices is usually achieved with quantum tomography. However, most methods which are currently widely used in experiments, such as maximum likelihood estimation, lack a well-justified error analysis. Promising recent methods based on confidence regions are difficult to apply in practice or yield error bars which are unnecessarily large. Here, we propose a practical yet robust method for obtaining error bars. We do so by introducing a novel representation of...
Classification Model for Forest Fire Hotspot Occurrences Prediction Using ANFIS Algorithm
Wijayanto, A. K.; Sani, O.; Kartika, N. D.; Herdiyeni, Y.
2017-01-01
This study proposed the application of data mining technique namely Adaptive Neuro-Fuzzy inference system (ANFIS) on forest fires hotspot data to develop classification models for hotspots occurrence in Central Kalimantan. Hotspot is a point that is indicated as the location of fires. In this study, hotspot distribution is categorized as true alarm and false alarm. ANFIS is a soft computing method in which a given inputoutput data set is expressed in a fuzzy inference system (FIS). The FIS implements a nonlinear mapping from its input space to the output space. The method of this study classified hotspots as target objects by correlating spatial attributes data using three folds in ANFIS algorithm to obtain the best model. The best result obtained from the 3rd fold provided low error for training (error = 0.0093676) and also low error testing result (error = 0.0093676). Attribute of distance to road is the most determining factor that influences the probability of true and false alarm where the level of human activities in this attribute is higher. This classification model can be used to develop early warning system of forest fire.
Butterfly Classification by HSI and RGB Color Models Using Neural Networks
Directory of Open Access Journals (Sweden)
Jorge E. Grajales-Múnera
2013-11-01
Full Text Available This study aims the classification of Butterfly species through the implementation of Neural Networks and Image Processing. A total of 9 species of Morpho genre which has blue as a characteristic color are processed. For Butterfly segmentation we used image processing tools such as: Binarization, edge processing and mathematical morphology. For data processing RGB values are obtained for every image which are converted to HSI color model to identify blue pixels and obtain the data to the proposed Neural Networks: Back-Propagation and Perceptron. For analysis and verification of results confusion matrix are built and analyzed with the results of neural networks with the lowest error levels. We obtain error levels close to 1% in classification of some Butterfly species.
Björkstén, Karin Sparring; Bergqvist, Monica; Andersén-Karlsson, Eva; Benson, Lina; Ulfvarson, Johanna
2016-08-24
Many studies address the prevalence of medication errors but few address medication errors serious enough to be regarded as malpractice. Other studies have analyzed the individual and system contributory factor leading to a medication error. Nurses have a key role in medication administration, and there are contradictory reports on the nurses' work experience in relation to the risk and type for medication errors. All medication errors where a nurse was held responsible for malpractice (n = 585) during 11 years in Sweden were included. A qualitative content analysis and classification according to the type and the individual and system contributory factors was made. In order to test for possible differences between nurses' work experience and associations within and between the errors and contributory factors, Fisher's exact test was used, and Cohen's kappa (k) was performed to estimate the magnitude and direction of the associations. There were a total of 613 medication errors in the 585 cases, the most common being "Wrong dose" (41 %), "Wrong patient" (13 %) and "Omission of drug" (12 %). In 95 % of the cases, an average of 1.4 individual contributory factors was found; the most common being "Negligence, forgetfulness or lack of attentiveness" (68 %), "Proper protocol not followed" (25 %), "Lack of knowledge" (13 %) and "Practice beyond scope" (12 %). In 78 % of the cases, an average of 1.7 system contributory factors was found; the most common being "Role overload" (36 %), "Unclear communication or orders" (30 %) and "Lack of adequate access to guidelines or unclear organisational routines" (30 %). The errors "Wrong patient due to mix-up of patients" and "Wrong route" and the contributory factors "Lack of knowledge" and "Negligence, forgetfulness or lack of attentiveness" were more common in less experienced nurses. The experienced nurses were more prone to "Practice beyond scope of practice" and to make errors in spite of "Lack of adequate
Vietnamese Document Representation and Classification
Nguyen, Giang-Son; Gao, Xiaoying; Andreae, Peter
Vietnamese is very different from English and little research has been done on Vietnamese document classification, or indeed, on any kind of Vietnamese language processing, and only a few small corpora are available for research. We created a large Vietnamese text corpus with about 18000 documents, and manually classified them based on different criteria such as topics and styles, giving several classification tasks of different difficulty levels. This paper introduces a new syllable-based document representation at the morphological level of the language for efficient classification. We tested the representation on our corpus with different classification tasks using six classification algorithms and two feature selection techniques. Our experiments show that the new representation is effective for Vietnamese categorization, and suggest that best performance can be achieved using syllable-pair document representation, an SVM with a polynomial kernel as the learning algorithm, and using Information gain and an external dictionary for feature selection.
A Spectral-Texture Kernel-Based Classification Method for Hyperspectral Images
Directory of Open Access Journals (Sweden)
Yi Wang
2016-11-01
Full Text Available Classification of hyperspectral images always suffers from high dimensionality and very limited labeled samples. Recently, the spectral-spatial classification has attracted considerable attention and can achieve higher classification accuracy and smoother classification maps. In this paper, a novel spectral-spatial classification method for hyperspectral images by using kernel methods is investigated. For a given hyperspectral image, the principle component analysis (PCA transform is first performed. Then, the first principle component of the input image is segmented into non-overlapping homogeneous regions by using the entropy rate superpixel (ERS algorithm. Next, the local spectral histogram model is applied to each homogeneous region to obtain the corresponding texture features. Because this step is performed within each homogenous region, instead of within a fixed-size image window, the obtained local texture features in the image are more accurate, which can effectively benefit the improvement of classification accuracy. In the following step, a contextual spectral-texture kernel is constructed by combining spectral information in the image and the extracted texture information using the linearity property of the kernel methods. Finally, the classification map is achieved by the support vector machines (SVM classifier using the proposed spectral-texture kernel. Experiments on two benchmark airborne hyperspectral datasets demonstrate that our method can effectively improve classification accuracies, even though only a very limited training sample is available. Specifically, our method can achieve from 8.26% to 15.1% higher in terms of overall accuracy than the traditional SVM classifier. The performance of our method was further compared to several state-of-the-art classification methods of hyperspectral images using objective quantitative measures and a visual qualitative evaluation.
Sreejith, Sreevarsha; Pereverzyev, Sergiy, Jr.; Kelvin, Lee S.; Marleau, Francine R.; Haltmeier, Markus; Ebner, Judith; Bland-Hawthorn, Joss; Driver, Simon P.; Graham, Alister W.; Holwerda, Benne W.; Hopkins, Andrew M.; Liske, Jochen; Loveday, Jon; Moffett, Amanda J.; Pimbblet, Kevin A.; Taylor, Edward N.; Wang, Lingyu; Wright, Angus H.
2018-03-01
We apply four statistical learning methods to a sample of 7941 galaxies (z test the feasibility of using automated algorithms to classify galaxies. Using 10 features measured for each galaxy (sizes, colours, shape parameters, and stellar mass), we apply the techniques of Support Vector Machines, Classification Trees, Classification Trees with Random Forest (CTRF) and Neural Networks, and returning True Prediction Ratios (TPRs) of 75.8 per cent, 69.0 per cent, 76.2 per cent, and 76.0 per cent, respectively. Those occasions whereby all four algorithms agree with each other yet disagree with the visual classification (`unanimous disagreement') serves as a potential indicator of human error in classification, occurring in ˜ 9 per cent of ellipticals, ˜ 9 per cent of little blue spheroids, ˜ 14 per cent of early-type spirals, ˜ 21 per cent of intermediate-type spirals, and ˜ 4 per cent of late-type spirals and irregulars. We observe that the choice of parameters rather than that of algorithms is more crucial in determining classification accuracy. Due to its simplicity in formulation and implementation, we recommend the CTRF algorithm for classifying future galaxy data sets. Adopting the CTRF algorithm, the TPRs of the five galaxy types are : E, 70.1 per cent; LBS, 75.6 per cent; S0-Sa, 63.6 per cent; Sab-Scd, 56.4 per cent, and Sd-Irr, 88.9 per cent. Further, we train a binary classifier using this CTRF algorithm that divides galaxies into spheroid-dominated (E, LBS, and S0-Sa) and disc-dominated (Sab-Scd and Sd-Irr), achieving an overall accuracy of 89.8 per cent. This translates into an accuracy of 84.9 per cent for spheroid-dominated systems and 92.5 per cent for disc-dominated systems.
Error Estimation for Indoor 802.11 Location Fingerprinting
DEFF Research Database (Denmark)
Lemelson, Hendrik; Kjærgaard, Mikkel Baun; Hansen, Rene
2009-01-01
providers could adapt their delivered services based on the estimated position error to achieve a higher service quality. Finally, system operators could use the information to inspect whether a location system provides satisfactory positioning accuracy throughout the covered area. For position error...
Machine Learning Techniques for Stellar Light Curve Classification
Hinners, Trisha A.; Tat, Kevin; Thorp, Rachel
2018-07-01
We apply machine learning techniques in an attempt to predict and classify stellar properties from noisy and sparse time-series data. We preprocessed over 94 GB of Kepler light curves from the Mikulski Archive for Space Telescopes (MAST) to classify according to 10 distinct physical properties using both representation learning and feature engineering approaches. Studies using machine learning in the field have been primarily done on simulated data, making our study one of the first to use real light-curve data for machine learning approaches. We tuned our data using previous work with simulated data as a template and achieved mixed results between the two approaches. Representation learning using a long short-term memory recurrent neural network produced no successful predictions, but our work with feature engineering was successful for both classification and regression. In particular, we were able to achieve values for stellar density, stellar radius, and effective temperature with low error (∼2%–4%) and good accuracy (∼75%) for classifying the number of transits for a given star. The results show promise for improvement for both approaches upon using larger data sets with a larger minority class. This work has the potential to provide a foundation for future tools and techniques to aid in the analysis of astrophysical data.
An extension of the receiver operating characteristic curve and AUC-optimal classification.
Takenouchi, Takashi; Komori, Osamu; Eguchi, Shinto
2012-10-01
While most proposed methods for solving classification problems focus on minimization of the classification error rate, we are interested in the receiver operating characteristic (ROC) curve, which provides more information about classification performance than the error rate does. The area under the ROC curve (AUC) is a natural measure for overall assessment of a classifier based on the ROC curve. We discuss a class of concave functions for AUC maximization in which a boosting-type algorithm including RankBoost is considered, and the Bayesian risk consistency and the lower bound of the optimum function are discussed. A procedure derived by maximizing a specific optimum function has high robustness, based on gross error sensitivity. Additionally, we focus on the partial AUC, which is the partial area under the ROC curve. For example, in medical screening, a high true-positive rate to the fixed lower false-positive rate is preferable and thus the partial AUC corresponding to lower false-positive rates is much more important than the remaining AUC. We extend the class of concave optimum functions for partial AUC optimality with the boosting algorithm. We investigated the validity of the proposed method through several experiments with data sets in the UCI repository.
Evaluation of drug administration errors in a teaching hospital
Directory of Open Access Journals (Sweden)
Berdot Sarah
2012-03-01
Full Text Available Abstract Background Medication errors can occur at any of the three steps of the medication use process: prescribing, dispensing and administration. We aimed to determine the incidence, type and clinical importance of drug administration errors and to identify risk factors. Methods Prospective study based on disguised observation technique in four wards in a teaching hospital in Paris, France (800 beds. A pharmacist accompanied nurses and witnessed the preparation and administration of drugs to all patients during the three drug rounds on each of six days per ward. Main outcomes were number, type and clinical importance of errors and associated risk factors. Drug administration error rate was calculated with and without wrong time errors. Relationship between the occurrence of errors and potential risk factors were investigated using logistic regression models with random effects. Results Twenty-eight nurses caring for 108 patients were observed. Among 1501 opportunities for error, 415 administrations (430 errors with one or more errors were detected (27.6%. There were 312 wrong time errors, ten simultaneously with another type of error, resulting in an error rate without wrong time error of 7.5% (113/1501. The most frequently administered drugs were the cardiovascular drugs (425/1501, 28.3%. The highest risks of error in a drug administration were for dermatological drugs. No potentially life-threatening errors were witnessed and 6% of errors were classified as having a serious or significant impact on patients (mainly omission. In multivariate analysis, the occurrence of errors was associated with drug administration route, drug classification (ATC and the number of patient under the nurse's care. Conclusion Medication administration errors are frequent. The identification of its determinants helps to undertake designed interventions.
Video genre classification using multimodal features
Jin, Sung Ho; Bae, Tae Meon; Choo, Jin Ho; Ro, Yong Man
2003-12-01
We propose a video genre classification method using multimodal features. The proposed method is applied for the preprocessing of automatic video summarization or the retrieval and classification of broadcasting video contents. Through a statistical analysis of low-level and middle-level audio-visual features in video, the proposed method can achieve good performance in classifying several broadcasting genres such as cartoon, drama, music video, news, and sports. In this paper, we adopt MPEG-7 audio-visual descriptors as multimodal features of video contents and evaluate the performance of the classification by feeding the features into a decision tree-based classifier which is trained by CART. The experimental results show that the proposed method can recognize several broadcasting video genres with a high accuracy and the classification performance with multimodal features is superior to the one with unimodal features in the genre classification.
Montesano, P. M.; Cook, B. D.; Sun, G.; Simard, M.; Zhang, Z.; Nelson, R. F.; Ranson, K. J.; Lutchke, S.; Blair, J. B.
2012-01-01
The synergistic use of active and passive remote sensing (i.e., data fusion) demonstrates the ability of spaceborne light detection and ranging (LiDAR), synthetic aperture radar (SAR) and multispectral imagery for achieving the accuracy requirements of a global forest biomass mapping mission. This data fusion approach also provides a means to extend 3D information from discrete spaceborne LiDAR measurements of forest structure across scales much larger than that of the LiDAR footprint. For estimating biomass, these measurements mix a number of errors including those associated with LiDAR footprint sampling over regional - global extents. A general framework for mapping above ground live forest biomass (AGB) with a data fusion approach is presented and verified using data from NASA field campaigns near Howland, ME, USA, to assess AGB and LiDAR sampling errors across a regionally representative landscape. We combined SAR and Landsat-derived optical (passive optical) image data to identify forest patches, and used image and simulated spaceborne LiDAR data to compute AGB and estimate LiDAR sampling error for forest patches and 100m, 250m, 500m, and 1km grid cells. Forest patches were delineated with Landsat-derived data and airborne SAR imagery, and simulated spaceborne LiDAR (SSL) data were derived from orbit and cloud cover simulations and airborne data from NASA's Laser Vegetation Imaging Sensor (L VIS). At both the patch and grid scales, we evaluated differences in AGB estimation and sampling error from the combined use of LiDAR with both SAR and passive optical and with either SAR or passive optical alone. This data fusion approach demonstrates that incorporating forest patches into the AGB mapping framework can provide sub-grid forest information for coarser grid-level AGB reporting, and that combining simulated spaceborne LiDAR with SAR and passive optical data are most useful for estimating AGB when measurements from LiDAR are limited because they minimized
Death Certification Errors and the Effect on Mortality Statistics.
McGivern, Lauri; Shulman, Leanne; Carney, Jan K; Shapiro, Steven; Bundock, Elizabeth
Errors in cause and manner of death on death certificates are common and affect families, mortality statistics, and public health research. The primary objective of this study was to characterize errors in the cause and manner of death on death certificates completed by non-Medical Examiners. A secondary objective was to determine the effects of errors on national mortality statistics. We retrospectively compared 601 death certificates completed between July 1, 2015, and January 31, 2016, from the Vermont Electronic Death Registration System with clinical summaries from medical records. Medical Examiners, blinded to original certificates, reviewed summaries, generated mock certificates, and compared mock certificates with original certificates. They then graded errors using a scale from 1 to 4 (higher numbers indicated increased impact on interpretation of the cause) to determine the prevalence of minor and major errors. They also compared International Classification of Diseases, 10th Revision (ICD-10) codes on original certificates with those on mock certificates. Of 601 original death certificates, 319 (53%) had errors; 305 (51%) had major errors; and 59 (10%) had minor errors. We found no significant differences by certifier type (physician vs nonphysician). We did find significant differences in major errors in place of death ( P statistics. Surveillance and certifier education must expand beyond local and state efforts. Simplifying and standardizing underlying literal text for cause of death may improve accuracy, decrease coding errors, and improve national mortality statistics.
Directory of Open Access Journals (Sweden)
Asriyanti Indah Pratiwi
2018-01-01
Full Text Available Sentiment analysis in a movie review is the needs of today lifestyle. Unfortunately, enormous features make the sentiment of analysis slow and less sensitive. Finding the optimum feature selection and classification is still a challenge. In order to handle an enormous number of features and provide better sentiment classification, an information-based feature selection and classification are proposed. The proposed method reduces more than 90% unnecessary features while the proposed classification scheme achieves 96% accuracy of sentiment classification. From the experimental results, it can be concluded that the combination of proposed feature selection and classification achieves the best performance so far.
Schmalz, M.; Ritter, G.
Accurate multispectral or hyperspectral signature classification is key to the nonimaging detection and recognition of space objects. Additionally, signature classification accuracy depends on accurate spectral endmember determination [1]. Previous approaches to endmember computation and signature classification were based on linear operators or neural networks (NNs) expressed in terms of the algebra (R, +, x) [1,2]. Unfortunately, class separation in these methods tends to be suboptimal, and the number of signatures that can be accurately classified often depends linearly on the number of NN inputs. This can lead to poor endmember distinction, as well as potentially significant classification errors in the presence of noise or densely interleaved signatures. In contrast to traditional CNNs, autoassociative morphological memories (AMM) are a construct similar to Hopfield autoassociatived memories defined on the (R, +, ?,?) lattice algebra [3]. Unlimited storage and perfect recall of noiseless real valued patterns has been proven for AMMs [4]. However, AMMs suffer from sensitivity to specific noise models, that can be characterized as erosive and dilative noise. On the other hand, the prior definition of a set of endmembers corresponds to material spectra lying on vertices of the minimum convex region covering the image data. These vertices can be characterized as morphologically independent patterns. It has further been shown that AMMs can be based on dendritic computation [3,6]. These techniques yield improved accuracy and class segmentation/separation ability in the presence of highly interleaved signature data. In this paper, we present a procedure for endmember determination based on AMM noise sensitivity, which employs morphological dendritic computation. We show that detected endmembers can be exploited by AMM based classification techniques, to achieve accurate signature classification in the presence of noise, closely spaced or interleaved signatures, and
Directory of Open Access Journals (Sweden)
Shaohong Tian
2016-11-01
Full Text Available The wetland classification from remotely sensed data is usually difficult due to the extensive seasonal vegetation dynamics and hydrological fluctuation. This study presents a random forest classification approach for the retrieval of the wetland landcover in the arid regions by fusing the Pléiade-1B data with multi-date Landsat-8 data. The segmentation of the Pléiade-1B multispectral image data was performed based on an object-oriented approach, and the geometric and spectral features were extracted for the segmented image objects. The normalized difference vegetation index (NDVI series data were also calculated from the multi-date Landsat-8 data, reflecting vegetation phenological changes in its growth cycle. The feature set extracted from the two sensors data was optimized and employed to create the random forest model for the classification of the wetland landcovers in the Ertix River in northern Xinjiang, China. Comparison with other classification methods such as support vector machine and artificial neural network classifiers indicates that the random forest classifier can achieve accurate classification with an overall accuracy of 93% and the Kappa coefficient of 0.92. The classification accuracy of the farming lands and water bodies that have distinct boundaries with the surrounding land covers was improved 5%–10% by making use of the property of geometric shapes. To remove the difficulty in the classification that was caused by the similar spectral features of the vegetation covers, the phenological difference and the textural information of co-occurrence gray matrix were incorporated into the classification, and the main wetland vegetation covers in the study area were derived from the two sensors data. The inclusion of phenological information in the classification enables the classification errors being reduced down, and the overall accuracy was improved approximately 10%. The results show that the proposed random forest
Errors and mistakes in breast ultrasound diagnostics
Directory of Open Access Journals (Sweden)
Wiesław Jakubowski
2012-09-01
Full Text Available Sonomammography is often the first additional examination performed in the diagnostics of breast diseases. The development of ultrasound imaging techniques, particularly the introduction of high frequency transducers, matrix transducers, harmonic imaging and finally, elastography, influenced the improvement of breast disease diagnostics. Neverthe‑ less, as in each imaging method, there are errors and mistakes resulting from the techni‑ cal limitations of the method, breast anatomy (fibrous remodeling, insufficient sensitivity and, in particular, specificity. Errors in breast ultrasound diagnostics can be divided into impossible to be avoided and potentially possible to be reduced. In this article the most frequently made errors in ultrasound have been presented, including the ones caused by the presence of artifacts resulting from volumetric averaging in the near and far field, artifacts in cysts or in dilated lactiferous ducts (reverberations, comet tail artifacts, lateral beam artifacts, improper setting of general enhancement or time gain curve or range. Errors dependent on the examiner, resulting in the wrong BIRADS‑usg classification, are divided into negative and positive errors. The sources of these errors have been listed. The methods of minimization of the number of errors made have been discussed, includ‑ ing the ones related to the appropriate examination technique, taking into account data from case history and the use of the greatest possible number of additional options such as: harmonic imaging, color and power Doppler and elastography. In the article examples of errors resulting from the technical conditions of the method have been presented, and those dependent on the examiner which are related to the great diversity and variation of ultrasound images of pathological breast lesions.
Geometrical error calibration in reflective surface testing based on reverse Hartmann test
Gong, Zhidong; Wang, Daodang; Xu, Ping; Wang, Chao; Liang, Rongguang; Kong, Ming; Zhao, Jun; Mo, Linhai; Mo, Shuhui
2017-08-01
In the fringe-illumination deflectometry based on reverse-Hartmann-test configuration, ray tracing of the modeled testing system is performed to reconstruct the test surface error. Careful calibration of system geometry is required to achieve high testing accuracy. To realize the high-precision surface testing with reverse Hartmann test, a computer-aided geometrical error calibration method is proposed. The aberrations corresponding to various geometrical errors are studied. With the aberration weights for various geometrical errors, the computer-aided optimization of system geometry with iterative ray tracing is carried out to calibration the geometrical error, and the accuracy in the order of subnanometer is achieved.
Morten Moshagen; Benjamin E. Hilbig
2011-01-01
Taking a falsificationist perspective, the present paper identifies two major shortcomings of existing approaches to comparative model evaluations in general and strategy classifications in particular. These are (1) failure to consider systematic error and (2) neglect of global model fit. Using adherence measures to evaluate competing models implicitly makes the unrealistic assumption that the error associated with the model predictions is entirely random. By means of simple schematic example...
Development of an intelligent system for ultrasonic flaw classification in weldments
International Nuclear Information System (INIS)
Song, Sung-Jin; Kim, Hak-Joon; Cho, Hyeon
2002-01-01
Even though ultrasonic pattern recognition is considered as the most effective and promising approach to flaw classification in weldments, its application to the realistic field inspection is still very limited due to the crucial barriers in cost, time and reliability. To reduce such barriers, previously we have proposed an intelligent system approach that consisted of the following four ingredients: (1) a PC-based ultrasonic testing (PC-UT) system; (2) an effective invariant ultrasonic flaw classification algorithm; (3) an intelligent flaw classification software; and (4) a database with abundant experimental flaw signals. In the present work, for performing the ultrasonic flaw classification in weldments in a real-time fashion in many real word situations, we develop an intelligent system, which is called the 'Intelligent Ultrasonic Evaluation System (IUES)' by the integration of the above four ingredients into a single, unified system. In addition, for the improvement of classification accuracy of flaws, especially slag inclusions, we expand the feature set by adding new informative features, and demonstrate the enhanced performance of the IUES with flaw signals in the database constructed previously. And then, to take care of the increased redundancy in the feature set due to the addition of features, we also propose two efficient schemes for feature selection: the forward selection with trial and error, and the forward selection with criteria of the error probability and the linear correlation coefficients of individual features
An extension of PPLS-DA for classification and comparison to ordinary PLS-DA.
Directory of Open Access Journals (Sweden)
Anna Telaar
Full Text Available Classification studies are widely applied, e.g. in biomedical research to classify objects/patients into predefined groups. The goal is to find a classification function/rule which assigns each object/patient to a unique group with the greatest possible accuracy (classification error. Especially in gene expression experiments often a lot of variables (genes are measured for only few objects/patients. A suitable approach is the well-known method PLS-DA, which searches for a transformation to a lower dimensional space. Resulting new components are linear combinations of the original variables. An advancement of PLS-DA leads to PPLS-DA, introducing a so called 'power parameter', which is maximized towards the correlation between the components and the group-membership. We introduce an extension of PPLS-DA for optimizing this power parameter towards the final aim, namely towards a minimal classification error. We compare this new extension with the original PPLS-DA and also with the ordinary PLS-DA using simulated and experimental datasets. For the investigated data sets with weak linear dependency between features/variables, no improvement is shown for PPLS-DA and for the extensions compared to PLS-DA. A very weak linear dependency, a low proportion of differentially expressed genes for simulated data, does not lead to an improvement of PPLS-DA over PLS-DA, but our extension shows a lower prediction error. On the contrary, for the data set with strong between-feature collinearity and a low proportion of differentially expressed genes and a large total number of genes, the prediction error of PPLS-DA and the extensions is clearly lower than for PLS-DA. Moreover we compare these prediction results with results of support vector machines with linear kernel and linear discriminant analysis.
Directory of Open Access Journals (Sweden)
Bruni Vanida
2010-01-01
Full Text Available Abstract Background Drug prescribing errors are frequent in the hospital setting and pharmacists play an important role in detection of these errors. The objectives of this study are (1 to describe the drug prescribing errors rate during the patient's stay, (2 to find which characteristics for a prescribing error are the most predictive of their reproduction the next day despite pharmacist's alert (i.e. override the alert. Methods We prospectively collected all medication order lines and prescribing errors during 18 days in 7 medical wards' using computerized physician order entry. We described and modelled the errors rate according to the chronology of hospital stay. We performed a classification and regression tree analysis to find which characteristics of alerts were predictive of their overriding (i.e. prescribing error repeated. Results 12 533 order lines were reviewed, 117 errors (errors rate 0.9% were observed and 51% of these errors occurred on the first day of the hospital stay. The risk of a prescribing error decreased over time. 52% of the alerts were overridden (i.e error uncorrected by prescribers on the following day. Drug omissions were the most frequently taken into account by prescribers. The classification and regression tree analysis showed that overriding pharmacist's alerts is first related to the ward of the prescriber and then to either Anatomical Therapeutic Chemical class of the drug or the type of error. Conclusions Since 51% of prescribing errors occurred on the first day of stay, pharmacist should concentrate his analysis of drug prescriptions on this day. The difference of overriding behavior between wards and according drug Anatomical Therapeutic Chemical class or type of error could also guide the validation tasks and programming of electronic alerts.
PERANCANGAN COMPUTER AIDED SYSTEM DALAM MENGANALISA HUMAN ERROR DI PERKERETAAPIAN INDONESIA
Directory of Open Access Journals (Sweden)
Wiwik Budiawan
2013-06-01
Full Text Available Kecelakaan kereta api (KA yang terjadi secara beruntun di Indonesia sudah berada pada tingkat kritis. Berdasarkan data dari Direktorat Jendral Perkeretaapian, dalam kurun 5 tahun terakhir (2005-2009 total terdapat 611 kecelakaan KA. Banyak faktor yang berkontribusi menyebabkan terjadinya kecelakaan, antara lain: sarana, prasarana, SDM operator (human error, eksternal, dan alam. Kegagalan manusia (Human error merupakan salah satu faktor yang berpotensi menyebabkan terjadinya suatu kecelakaan KA dan dinyatakan sebagai faktor utama penyebab terjadinya suatu kecelakaan kereta api di Indonesia. Namun, tidak jelas bagaimana teknik analisis ini dilakukan. Kajian human error yang dilakukan Komite Nasional Keselamatan Transportasi (KNKT masih relatif terbatas, tidak dilengkapi dengan metode yang sistematis. Terdapat beberapa metode yang telah dikembangkan saat ini, tetapi untuk moda transportasi kereta api masih belum banyak dikembangkan. Human Factors Analysis and Classification System (HFACS merupakan metode analisis human error yang dikembangkan dan disesuaikan dengan sistem perkeretaapian Indonesia. Guna meningkatkan keandalan dalam analisis human error, HFACS kemudian dikembangkan dalam bentuk aplikasi berbasis web yang dapat diakses di komputer maupun smartphone. Hasil penelitian ini dapat dimanfaatkan oleh KNKT sebagai metode analisis kecelakaan kereta api khususnya terkait dengan human error. Kata kunci : human error, HFACS, CAS, kereta api Abstract Train wreck (KA which occurred in quick succession in Indonesia already at a critical level. Based on data from the Directorate General of Railways, during the last 5 years (2005-2009 there were a total of 611 railway accidents. Many factors contribute to cause accidents, such as: facilities, infrastructure, human operator (human error, external, and natural. Human failure (Human error is one of the factors that could potentially cause a train accident and expressed as the main factors causing
Directory of Open Access Journals (Sweden)
G. A. Shirali
2013-12-01
.Result: In this study, 158 reports of accident in Ahvaz steel industry were analyzed by HFACS technique. This analysis showed that most of the human errors were: in the first level was related to the skill-based errors, in the second to the physical environment, in the third level to the inadequate supervision and in the fourth level to the management of resources. .Conclusion: Studying and analyzing of past events using the HFACS technique can identify the major and root causes of accidents and can be effective on prevent repetitions of such mishaps. Also, it can be used as a basis for developing strategies to prevent future events in steel industries.
Discriminative Hierarchical K-Means Tree for Large-Scale Image Classification.
Chen, Shizhi; Yang, Xiaodong; Tian, Yingli
2015-09-01
A key challenge in large-scale image classification is how to achieve efficiency in terms of both computation and memory without compromising classification accuracy. The learning-based classifiers achieve the state-of-the-art accuracies, but have been criticized for the computational complexity that grows linearly with the number of classes. The nonparametric nearest neighbor (NN)-based classifiers naturally handle large numbers of categories, but incur prohibitively expensive computation and memory costs. In this brief, we present a novel classification scheme, i.e., discriminative hierarchical K-means tree (D-HKTree), which combines the advantages of both learning-based and NN-based classifiers. The complexity of the D-HKTree only grows sublinearly with the number of categories, which is much better than the recent hierarchical support vector machines-based methods. The memory requirement is the order of magnitude less than the recent Naïve Bayesian NN-based approaches. The proposed D-HKTree classification scheme is evaluated on several challenging benchmark databases and achieves the state-of-the-art accuracies, while with significantly lower computation cost and memory requirement.
International Nuclear Information System (INIS)
Trehan, Sumeet; Carlberg, Kevin T.; Durlofsky, Louis J.
2017-01-01
A machine learning–based framework for modeling the error introduced by surrogate models of parameterized dynamical systems is proposed. The framework entails the use of high-dimensional regression techniques (eg, random forests, and LASSO) to map a large set of inexpensively computed “error indicators” (ie, features) produced by the surrogate model at a given time instance to a prediction of the surrogate-model error in a quantity of interest (QoI). This eliminates the need for the user to hand-select a small number of informative features. The methodology requires a training set of parameter instances at which the time-dependent surrogate-model error is computed by simulating both the high-fidelity and surrogate models. Using these training data, the method first determines regression-model locality (via classification or clustering) and subsequently constructs a “local” regression model to predict the time-instantaneous error within each identified region of feature space. We consider 2 uses for the resulting error model: (1) as a correction to the surrogate-model QoI prediction at each time instance and (2) as a way to statistically model arbitrary functions of the time-dependent surrogate-model error (eg, time-integrated errors). We then apply the proposed framework to model errors in reduced-order models of nonlinear oil-water subsurface flow simulations, with time-varying well-control (bottom-hole pressure) parameters. The reduced-order models used in this work entail application of trajectory piecewise linearization in conjunction with proper orthogonal decomposition. Moreover, when the first use of the method is considered, numerical experiments demonstrate consistent improvement in accuracy in the time-instantaneous QoI prediction relative to the original surrogate model, across a large number of test cases. When the second use is considered, results show that the proposed method provides accurate statistical predictions of the time- and well
Spelling in Adolescents with Dyslexia: Errors and Modes of Assessment
Tops, Wim; Callens, Maaike; Bijn, Evi; Brysbaert, Marc
2014-01-01
In this study we focused on the spelling of high-functioning students with dyslexia. We made a detailed classification of the errors in a word and sentence dictation task made by 100 students with dyslexia and 100 matched control students. All participants were in the first year of their bachelor's studies and had Dutch as mother tongue. Three…
Identifying medication error chains from critical incident reports: a new analytic approach.
Huckels-Baumgart, Saskia; Manser, Tanja
2014-10-01
Research into the distribution of medication errors usually focuses on isolated stages within the medication use process. Our study aimed to provide a novel process-oriented approach to medication incident analysis focusing on medication error chains. Our study was conducted across a 900-bed teaching hospital in Switzerland. All reported 1,591 medication errors 2009-2012 were categorized using the Medication Error Index NCC MERP and the WHO Classification for Patient Safety Methodology. In order to identify medication error chains, each reported medication incident was allocated to the relevant stage of the hospital medication use process. Only 25.8% of the reported medication errors were detected before they propagated through the medication use process. The majority of medication errors (74.2%) formed an error chain encompassing two or more stages. The most frequent error chain comprised preparation up to and including medication administration (45.2%). "Non-consideration of documentation/prescribing" during the drug preparation was the most frequent contributor for "wrong dose" during the administration of medication. Medication error chains provide important insights for detecting and stopping medication errors before they reach the patient. Existing and new safety barriers need to be extended to interrupt error chains and to improve patient safety. © 2014, The American College of Clinical Pharmacology.
Spüler, Martin; Niethammer, Christian
2015-01-01
When a person recognizes an error during a task, an error-related potential (ErrP) can be measured as response. It has been shown that ErrPs can be automatically detected in tasks with time-discrete feedback, which is widely applied in the field of Brain-Computer Interfaces (BCIs) for error correction or adaptation. However, there are only a few studies that concentrate on ErrPs during continuous feedback. With this study, we wanted to answer three different questions: (i) Can ErrPs be measured in electroencephalography (EEG) recordings during a task with continuous cursor control? (ii) Can ErrPs be classified using machine learning methods and is it possible to discriminate errors of different origins? (iii) Can we use EEG to detect the severity of an error? To answer these questions, we recorded EEG data from 10 subjects during a video game task and investigated two different types of error (execution error, due to inaccurate feedback; outcome error, due to not achieving the goal of an action). We analyzed the recorded data to show that during the same task, different kinds of error produce different ErrP waveforms and have a different spectral response. This allows us to detect and discriminate errors of different origin in an event-locked manner. By utilizing the error-related spectral response, we show that also a continuous, asynchronous detection of errors is possible. Although the detection of error severity based on EEG was one goal of this study, we did not find any significant influence of the severity on the EEG. PMID:25859204
Directory of Open Access Journals (Sweden)
Martin eSpüler
2015-03-01
Full Text Available When a person recognizes an error during a task, an error-related potential (ErrP can be measured as response. It has been shown that ErrPs can be automatically detected in tasks with time-discrete feedback, which is widely applied in the field of Brain-Computer Interfaces (BCIs for error correction or adaptation. However, there are only a few studies that concentrate on ErrPs during continuous feedback.With this study, we wanted to answer three different questions: (i Can ErrPs be measured in electroencephalography (EEG recordings during a task with continuous cursor control? (ii Can ErrPs be classified using machine learning methods and is it possible to discriminate errors of different origins? (iii Can we use EEG to detect the severity of an error? To answer these questions, we recorded EEG data from 10 subjects during a video game task and investigated two different types of error (execution error, due to inaccurate feedback; outcome error, due to not achieving the goal of an action. We analyzed the recorded data to show that during the same task, different kinds of error produce different ErrP waveforms and have a different spectral response. This allows us to detect and discriminate errors of different origin in an event-locked manner. By utilizing the error-related spectral response, we show that also a continuous, asynchronous detection of errors is possible.Although the detection of error severity based on EEG was one goal of this study, we did not find any significant influence of the severity on the EEG.
Directory of Open Access Journals (Sweden)
Mark D McDonnell
Full Text Available Recent advances in training deep (multi-layer architectures have inspired a renaissance in neural network use. For example, deep convolutional networks are becoming the default option for difficult tasks on large datasets, such as image and speech recognition. However, here we show that error rates below 1% on the MNIST handwritten digit benchmark can be replicated with shallow non-convolutional neural networks. This is achieved by training such networks using the 'Extreme Learning Machine' (ELM approach, which also enables a very rapid training time (∼ 10 minutes. Adding distortions, as is common practise for MNIST, reduces error rates even further. Our methods are also shown to be capable of achieving less than 5.5% error rates on the NORB image database. To achieve these results, we introduce several enhancements to the standard ELM algorithm, which individually and in combination can significantly improve performance. The main innovation is to ensure each hidden-unit operates only on a randomly sized and positioned patch of each image. This form of random 'receptive field' sampling of the input ensures the input weight matrix is sparse, with about 90% of weights equal to zero. Furthermore, combining our methods with a small number of iterations of a single-batch backpropagation method can significantly reduce the number of hidden-units required to achieve a particular performance. Our close to state-of-the-art results for MNIST and NORB suggest that the ease of use and accuracy of the ELM algorithm for designing a single-hidden-layer neural network classifier should cause it to be given greater consideration either as a standalone method for simpler problems, or as the final classification stage in deep neural networks applied to more difficult problems.
Errors generated with the use of rectangular collimation
International Nuclear Information System (INIS)
Parks, E.T.
1991-01-01
This study was designed to determine whether various techniques for achieving rectangular collimation generate different numbers and types of errors and remakes and to determine whether operator skill level influences errors and remakes. Eighteen students exposed full-mouth series of radiographs on manikins with the use of six techniques. The students were grouped according to skill level. The radiographs were evaluated for errors and remakes resulting from errors in the following categories: cone cutting, vertical angulation, and film placement. Significant differences were found among the techniques in cone cutting errors and remakes, vertical angulation errors and remakes, and total errors and remakes. Operator skill did not appear to influence the number or types of errors or remakes generated. Rectangular collimation techniques produced more errors than did the round collimation techniques. However, only one rectangular collimation technique generated significantly more remakes than the other techniques
The VTTVIS line imaging spectrometer - principles, error sources, and calibration
DEFF Research Database (Denmark)
Jørgensen, R.N.
2002-01-01
work describing the basic principles, potential error sources, and/or adjustment and calibration procedures. This report fulfils the need for such documentationwith special focus on the system at KVL. The PGP based system has several severe error sources, which should be removed prior any analysis......Hyperspectral imaging with a spatial resolution of a few mm2 has proved to have a great potential within crop and weed classification and also within nutrient diagnostics. A commonly used hyperspectral imaging system is based on the Prism-Grating-Prism(PGP) principles produced by Specim Ltd...... in off-axis transmission efficiencies, diffractionefficiencies, and image distortion have a significant impact on the instrument performance. Procedures removing or minimising these systematic error sources are developed and described for the system build at KVL but can be generalised to other PGP...
Comparative analysis of methods for classification in predicting the quality of bread
E. A. Balashova; V. K. Bitjukov; E. A. Savvina
2013-01-01
The comparative analysis of classification methods of two-stage cluster and discriminant analysis and neural networks was performed. System of informative signs which classifies with a minimum of errors has been proposed.
Hyperspectral image classification based on local binary patterns and PCANet
Yang, Huizhen; Gao, Feng; Dong, Junyu; Yang, Yang
2018-04-01
Hyperspectral image classification has been well acknowledged as one of the challenging tasks of hyperspectral data processing. In this paper, we propose a novel hyperspectral image classification framework based on local binary pattern (LBP) features and PCANet. In the proposed method, linear prediction error (LPE) is first employed to select a subset of informative bands, and LBP is utilized to extract texture features. Then, spectral and texture features are stacked into a high dimensional vectors. Next, the extracted features of a specified position are transformed to a 2-D image. The obtained images of all pixels are fed into PCANet for classification. Experimental results on real hyperspectral dataset demonstrate the effectiveness of the proposed method.
Refractive errors in children and adolescents in Bucaramanga (Colombia).
Galvis, Virgilio; Tello, Alejandro; Otero, Johanna; Serrano, Andrés A; Gómez, Luz María; Castellanos, Yuly
2017-01-01
The aim of this study was to establish the frequency of refractive errors in children and adolescents aged between 8 and 17 years old, living in the metropolitan area of Bucaramanga (Colombia). This study was a secondary analysis of two descriptive cross-sectional studies that applied sociodemographic surveys and assessed visual acuity and refraction. Ametropias were classified as myopic errors, hyperopic errors, and mixed astigmatism. Eyes were considered emmetropic if none of these classifications were made. The data were collated using free software and analyzed with STATA/IC 11.2. One thousand two hundred twenty-eight individuals were included in this study. Girls showed a higher rate of ametropia than boys. Hyperopic refractive errors were present in 23.1% of the subjects, and myopic errors in 11.2%. Only 0.2% of the eyes had high myopia (≤-6.00 D). Mixed astigmatism and anisometropia were uncommon, and myopia frequency increased with age. There were statistically significant steeper keratometric readings in myopic compared to hyperopic eyes. The frequency of refractive errors that we found of 36.7% is moderate compared to the global data. The rates and parameters statistically differed by sex and age groups. Our findings are useful for establishing refractive error rate benchmarks in low-middle-income countries and as a baseline for following their variation by sociodemographic factors.
Refractive errors in children and adolescents in Bucaramanga (Colombia
Directory of Open Access Journals (Sweden)
Virgilio Galvis
Full Text Available ABSTRACT Purpose: The aim of this study was to establish the frequency of refractive errors in children and adolescents aged between 8 and 17 years old, living in the metropolitan area of Bucaramanga (Colombia. Methods: This study was a secondary analysis of two descriptive cross-sectional studies that applied sociodemographic surveys and assessed visual acuity and refraction. Ametropias were classified as myopic errors, hyperopic errors, and mixed astigmatism. Eyes were considered emmetropic if none of these classifications were made. The data were collated using free software and analyzed with STATA/IC 11.2. Results: One thousand two hundred twenty-eight individuals were included in this study. Girls showed a higher rate of ametropia than boys. Hyperopic refractive errors were present in 23.1% of the subjects, and myopic errors in 11.2%. Only 0.2% of the eyes had high myopia (≤-6.00 D. Mixed astigmatism and anisometropia were uncommon, and myopia frequency increased with age. There were statistically significant steeper keratometric readings in myopic compared to hyperopic eyes. Conclusions: The frequency of refractive errors that we found of 36.7% is moderate compared to the global data. The rates and parameters statistically differed by sex and age groups. Our findings are useful for establishing refractive error rate benchmarks in low-middle-income countries and as a baseline for following their variation by sociodemographic factors.
Constrained motion estimation-based error resilient coding for HEVC
Guo, Weihan; Zhang, Yongfei; Li, Bo
2018-04-01
Unreliable communication channels might lead to packet losses and bit errors in the videos transmitted through it, which will cause severe video quality degradation. This is even worse for HEVC since more advanced and powerful motion estimation methods are introduced to further remove the inter-frame dependency and thus improve the coding efficiency. Once a Motion Vector (MV) is lost or corrupted, it will cause distortion in the decoded frame. More importantly, due to motion compensation, the error will propagate along the motion prediction path, accumulate over time, and significantly degrade the overall video presentation quality. To address this problem, we study the problem of encoder-sider error resilient coding for HEVC and propose a constrained motion estimation scheme to mitigate the problem of error propagation to subsequent frames. The approach is achieved by cutting off MV dependencies and limiting the block regions which are predicted by temporal motion vector. The experimental results show that the proposed method can effectively suppress the error propagation caused by bit errors of motion vector and can improve the robustness of the stream in the bit error channels. When the bit error probability is 10-5, an increase of the decoded video quality (PSNR) by up to1.310dB and on average 0.762 dB can be achieved, compared to the reference HEVC.
A novel Neuro-fuzzy classification technique for data mining
Directory of Open Access Journals (Sweden)
Soumadip Ghosh
2014-11-01
Full Text Available In our study, we proposed a novel Neuro-fuzzy classification technique for data mining. The inputs to the Neuro-fuzzy classification system were fuzzified by applying generalized bell-shaped membership function. The proposed method utilized a fuzzification matrix in which the input patterns were associated with a degree of membership to different classes. Based on the value of degree of membership a pattern would be attributed to a specific category or class. We applied our method to ten benchmark data sets from the UCI machine learning repository for classification. Our objective was to analyze the proposed method and, therefore compare its performance with two powerful supervised classification algorithms Radial Basis Function Neural Network (RBFNN and Adaptive Neuro-fuzzy Inference System (ANFIS. We assessed the performance of these classification methods in terms of different performance measures such as accuracy, root-mean-square error, kappa statistic, true positive rate, false positive rate, precision, recall, and f-measure. In every aspect the proposed method proved to be superior to RBFNN and ANFIS algorithms.
Neuromuscular disease classification system
Sáez, Aurora; Acha, Begoña; Montero-Sánchez, Adoración; Rivas, Eloy; Escudero, Luis M.; Serrano, Carmen
2013-06-01
Diagnosis of neuromuscular diseases is based on subjective visual assessment of biopsies from patients by the pathologist specialist. A system for objective analysis and classification of muscular dystrophies and neurogenic atrophies through muscle biopsy images of fluorescence microscopy is presented. The procedure starts with an accurate segmentation of the muscle fibers using mathematical morphology and a watershed transform. A feature extraction step is carried out in two parts: 24 features that pathologists take into account to diagnose the diseases and 58 structural features that the human eye cannot see, based on the assumption that the biopsy is considered as a graph, where the nodes are represented by each fiber, and two nodes are connected if two fibers are adjacent. A feature selection using sequential forward selection and sequential backward selection methods, a classification using a Fuzzy ARTMAP neural network, and a study of grading the severity are performed on these two sets of features. A database consisting of 91 images was used: 71 images for the training step and 20 as the test. A classification error of 0% was obtained. It is concluded that the addition of features undetectable by the human visual inspection improves the categorization of atrophic patterns.
Unsupervised classification of variable stars
Valenzuela, Lucas; Pichara, Karim
2018-03-01
During the past 10 years, a considerable amount of effort has been made to develop algorithms for automatic classification of variable stars. That has been primarily achieved by applying machine learning methods to photometric data sets where objects are represented as light curves. Classifiers require training sets to learn the underlying patterns that allow the separation among classes. Unfortunately, building training sets is an expensive process that demands a lot of human efforts. Every time data come from new surveys; the only available training instances are the ones that have a cross-match with previously labelled objects, consequently generating insufficient training sets compared with the large amounts of unlabelled sources. In this work, we present an algorithm that performs unsupervised classification of variable stars, relying only on the similarity among light curves. We tackle the unsupervised classification problem by proposing an untraditional approach. Instead of trying to match classes of stars with clusters found by a clustering algorithm, we propose a query-based method where astronomers can find groups of variable stars ranked by similarity. We also develop a fast similarity function specific for light curves, based on a novel data structure that allows scaling the search over the entire data set of unlabelled objects. Experiments show that our unsupervised model achieves high accuracy in the classification of different types of variable stars and that the proposed algorithm scales up to massive amounts of light curves.
KMRR thermal power measurement error estimation
International Nuclear Information System (INIS)
Rhee, B.W.; Sim, B.S.; Lim, I.C.; Oh, S.K.
1990-01-01
The thermal power measurement error of the Korea Multi-purpose Research Reactor has been estimated by a statistical Monte Carlo method, and compared with those obtained by the other methods including deterministic and statistical approaches. The results show that the specified thermal power measurement error of 5% cannot be achieved if the commercial RTDs are used to measure the coolant temperatures of the secondary cooling system and the error can be reduced below the requirement if the commercial RTDs are replaced by the precision RTDs. The possible range of the thermal power control operation has been identified to be from 100% to 20% of full power
Class-specific Error Bounds for Ensemble Classifiers
Energy Technology Data Exchange (ETDEWEB)
Prenger, R; Lemmond, T; Varshney, K; Chen, B; Hanley, W
2009-10-06
The generalization error, or probability of misclassification, of ensemble classifiers has been shown to be bounded above by a function of the mean correlation between the constituent (i.e., base) classifiers and their average strength. This bound suggests that increasing the strength and/or decreasing the correlation of an ensemble's base classifiers may yield improved performance under the assumption of equal error costs. However, this and other existing bounds do not directly address application spaces in which error costs are inherently unequal. For applications involving binary classification, Receiver Operating Characteristic (ROC) curves, performance curves that explicitly trade off false alarms and missed detections, are often utilized to support decision making. To address performance optimization in this context, we have developed a lower bound for the entire ROC curve that can be expressed in terms of the class-specific strength and correlation of the base classifiers. We present empirical analyses demonstrating the efficacy of these bounds in predicting relative classifier performance. In addition, we specify performance regions of the ROC curve that are naturally delineated by the class-specific strengths of the base classifiers and show that each of these regions can be associated with a unique set of guidelines for performance optimization of binary classifiers within unequal error cost regimes.
Bayesian Optimization for Neuroimaging Pre-processing in Brain Age Classification and Prediction
Directory of Open Access Journals (Sweden)
Jenessa Lancaster
2018-02-01
Full Text Available Neuroimaging-based age prediction using machine learning is proposed as a biomarker of brain aging, relating to cognitive performance, health outcomes and progression of neurodegenerative disease. However, even leading age-prediction algorithms contain measurement error, motivating efforts to improve experimental pipelines. T1-weighted MRI is commonly used for age prediction, and the pre-processing of these scans involves normalization to a common template and resampling to a common voxel size, followed by spatial smoothing. Resampling parameters are often selected arbitrarily. Here, we sought to improve brain-age prediction accuracy by optimizing resampling parameters using Bayesian optimization. Using data on N = 2003 healthy individuals (aged 16–90 years we trained support vector machines to (i distinguish between young (<22 years and old (>50 years brains (classification and (ii predict chronological age (regression. We also evaluated generalisability of the age-regression model to an independent dataset (CamCAN, N = 648, aged 18–88 years. Bayesian optimization was used to identify optimal voxel size and smoothing kernel size for each task. This procedure adaptively samples the parameter space to evaluate accuracy across a range of possible parameters, using independent sub-samples to iteratively assess different parameter combinations to arrive at optimal values. When distinguishing between young and old brains a classification accuracy of 88.1% was achieved, (optimal voxel size = 11.5 mm3, smoothing kernel = 2.3 mm. For predicting chronological age, a mean absolute error (MAE of 5.08 years was achieved, (optimal voxel size = 3.73 mm3, smoothing kernel = 3.68 mm. This was compared to performance using default values of 1.5 mm3 and 4mm respectively, resulting in MAE = 5.48 years, though this 7.3% improvement was not statistically significant. When assessing generalisability, best performance was achieved when applying the entire Bayesian
Quantitative assessment of errors in monitoring landcover changes by comparison of maps
Directory of Open Access Journals (Sweden)
Jean Francois Mas
2012-02-01
Full Text Available Many studies aimed at assessing land-cover changes are based upon the comparison of maps elaborated in different dates. This comparison allows the calculation of change rates as well as to generate more detailed data such as the transition matrix and the change map. In this study, we evaluated the errors incurred when comparing maps elaborated at different scales, obtained through independent digitalisation processes, elaborated using different classification schemes or when the maps were elaborated with inputs from different dates. Errors derived from the difference of scale or from the map-digitalisation processes led to false changes with a similar or greater scale to that of true changes. The comparison of maps based on different classification schemes invalidated the results of the comparison. By contrast, the different approaches used to tackle the issue of maps with multiple dates produced similar results. The paper discusses some methods aimed at reducing these problems and evaluating the reliability of multi-temporal databases.
Featureless classification of light curves
Kügler, S. D.; Gianniotis, N.; Polsterer, K. L.
2015-08-01
In the era of rapidly increasing amounts of time series data, classification of variable objects has become the main objective of time-domain astronomy. Classification of irregularly sampled time series is particularly difficult because the data cannot be represented naturally as a vector which can be directly fed into a classifier. In the literature, various statistical features serve as vector representations. In this work, we represent time series by a density model. The density model captures all the information available, including measurement errors. Hence, we view this model as a generalization to the static features which directly can be derived, e.g. as moments from the density. Similarity between each pair of time series is quantified by the distance between their respective models. Classification is performed on the obtained distance matrix. In the numerical experiments, we use data from the OGLE (Optical Gravitational Lensing Experiment) and ASAS (All Sky Automated Survey) surveys and demonstrate that the proposed representation performs up to par with the best currently used feature-based approaches. The density representation preserves all static information present in the observational data, in contrast to a less-complete description by features. The density representation is an upper boundary in terms of information made available to the classifier. Consequently, the predictive power of the proposed classification depends on the choice of similarity measure and classifier, only. Due to its principled nature, we advocate that this new approach of representing time series has potential in tasks beyond classification, e.g. unsupervised learning.
Analysis and application of classification methods of complex carbonate reservoirs
Li, Xiongyan; Qin, Ruibao; Ping, Haitao; Wei, Dan; Liu, Xiaomei
2018-06-01
There are abundant carbonate reservoirs from the Cenozoic to Mesozoic era in the Middle East. Due to variation in sedimentary environment and diagenetic process of carbonate reservoirs, several porosity types coexist in carbonate reservoirs. As a result, because of the complex lithologies and pore types as well as the impact of microfractures, the pore structure is very complicated. Therefore, it is difficult to accurately calculate the reservoir parameters. In order to accurately evaluate carbonate reservoirs, based on the pore structure evaluation of carbonate reservoirs, the classification methods of carbonate reservoirs are analyzed based on capillary pressure curves and flow units. Based on the capillary pressure curves, although the carbonate reservoirs can be classified, the relationship between porosity and permeability after classification is not ideal. On the basis of the flow units, the high-precision functional relationship between porosity and permeability after classification can be established. Therefore, the carbonate reservoirs can be quantitatively evaluated based on the classification of flow units. In the dolomite reservoirs, the average absolute error of calculated permeability decreases from 15.13 to 7.44 mD. Similarly, the average absolute error of calculated permeability of limestone reservoirs is reduced from 20.33 to 7.37 mD. Only by accurately characterizing pore structures and classifying reservoir types, reservoir parameters could be calculated accurately. Therefore, characterizing pore structures and classifying reservoir types are very important to accurate evaluation of complex carbonate reservoirs in the Middle East.
Classifier fusion for VoIP attacks classification
Safarik, Jakub; Rezac, Filip
2017-05-01
SIP is one of the most successful protocols in the field of IP telephony communication. It establishes and manages VoIP calls. As the number of SIP implementation rises, we can expect a higher number of attacks on the communication system in the near future. This work aims at malicious SIP traffic classification. A number of various machine learning algorithms have been developed for attack classification. The paper presents a comparison of current research and the use of classifier fusion method leading to a potential decrease in classification error rate. Use of classifier combination makes a more robust solution without difficulties that may affect single algorithms. Different voting schemes, combination rules, and classifiers are discussed to improve the overall performance. All classifiers have been trained on real malicious traffic. The concept of traffic monitoring depends on the network of honeypot nodes. These honeypots run in several networks spread in different locations. Separation of honeypots allows us to gain an independent and trustworthy attack information.
Error-Resilient Unequal Error Protection of Fine Granularity Scalable Video Bitstreams
Cai, Hua; Zeng, Bing; Shen, Guobin; Xiong, Zixiang; Li, Shipeng
2006-12-01
This paper deals with the optimal packet loss protection issue for streaming the fine granularity scalable (FGS) video bitstreams over IP networks. Unlike many other existing protection schemes, we develop an error-resilient unequal error protection (ER-UEP) method that adds redundant information optimally for loss protection and, at the same time, cancels completely the dependency among bitstream after loss recovery. In our ER-UEP method, the FGS enhancement-layer bitstream is first packetized into a group of independent and scalable data packets. Parity packets, which are also scalable, are then generated. Unequal protection is finally achieved by properly shaping the data packets and the parity packets. We present an algorithm that can optimally allocate the rate budget between data packets and parity packets, together with several simplified versions that have lower complexity. Compared with conventional UEP schemes that suffer from bit contamination (caused by the bit dependency within a bitstream), our method guarantees successful decoding of all received bits, thus leading to strong error-resilience (at any fixed channel bandwidth) and high robustness (under varying and/or unclean channel conditions).
Customer and performance rating in QFD using SVM classification
Dzulkifli, Syarizul Amri; Salleh, Mohd Najib Mohd; Leman, A. M.
2017-09-01
In a classification problem, where each input is associated to one output. Training data is used to create a model which predicts values to the true function. SVM is a popular method for binary classification due to their theoretical foundation and good generalization performance. However, when trained with noisy data, the decision hyperplane might deviate from optimal position because of the sum of misclassification errors in the objective function. In this paper, we introduce fuzzy in weighted learning approach for improving the accuracy of Support Vector Machine (SVM) classification. The main aim of this work is to determine appropriate weighted for SVM to adjust the parameters of learning method from a given set of noisy input to output data. The performance and customer rating in Quality Function Deployment (QFD) is used as our case study to determine implementing fuzzy SVM is highly scalable for very large data sets and generating high classification accuracy.
Research on Classification of Chinese Text Data Based on SVM
Lin, Yuan; Yu, Hongzhi; Wan, Fucheng; Xu, Tao
2017-09-01
Data Mining has important application value in today’s industry and academia. Text classification is a very important technology in data mining. At present, there are many mature algorithms for text classification. KNN, NB, AB, SVM, decision tree and other classification methods all show good classification performance. Support Vector Machine’ (SVM) classification method is a good classifier in machine learning research. This paper will study the classification effect based on the SVM method in the Chinese text data, and use the support vector machine method in the chinese text to achieve the classify chinese text, and to able to combination of academia and practical application.
Comparative analysis of methods for classification in predicting the quality of bread
Directory of Open Access Journals (Sweden)
E. A. Balashova
2013-01-01
Full Text Available The comparative analysis of classification methods of two-stage cluster and discriminant analysis and neural networks was performed. System of informative signs which classifies with a minimum of errors has been proposed.
Galaxy Zoo 1: data release of morphological classifications for nearly 900 000 galaxies
Energy Technology Data Exchange (ETDEWEB)
Linott, C.; Slosar, A.; Lintott, C.; Schawinski, K.; Bamford, S.; Slosar, A.; Land, K.; Thomas, D.; Edmondson, E.; Masters, K.; Nichol, R.C.; Raddick, M.J.; Szalay, A.; Andreescu, D.; Murray, P.; Vandenberg, J.
2011-01-01
Morphology is a powerful indicator of a galaxy's dynamical and merger history. It is strongly correlated with many physical parameters, including mass, star formation history and the distribution of mass. The Galaxy Zoo project collected simple morphological classifications of nearly 900,000 galaxies drawn from the Sloan Digital Sky Survey, contributed by hundreds of thousands of volunteers. This large number of classifications allows us to exclude classifier error, and measure the influence of subtle biases inherent in morphological classification. This paper presents the data collected by the project, alongside measures of classification accuracy and bias. The data are now publicly available and full catalogues can be downloaded in electronic format from http://data.galaxyzoo.org.
Directory of Open Access Journals (Sweden)
Ranauli Sihombing
2016-12-01
Full Text Available Errors analysis has become one of the most interesting issues in the study of Second Language Acquisition. It can not be denied that some teachers do not know a lot about error analysis and related theories of how L1, L2 or foreign language acquired. In addition, the students often feel upset since they find a gap between themselves and the teachers for the errors the students make and the teachers’ understanding about the error correction. The present research aims to investigate what errors adult English learners make in written production of English. The significances of the study is to know what errors students make in writing that the teachers can find solution to the errors the students make for a better English language teaching and learning especially in teaching English for adults. The study employed qualitative method. The research was undertaken at an airline education center in Bandung. The result showed that syntax errors are more frequently found than morphology errors, especially in terms of verb phrase errors. It is recommended that it is important for teacher to know the theory of second language acquisition in order to know how the students learn and produce theirlanguage. In addition, it will be advantages for teachers if they know what errors students frequently make in their learning, so that the teachers can give solution to the students for a better English language learning achievement. DOI: https://doi.org/10.24071/llt.2015.180205
Dieye, A.M.; Roy, David P.; Hanan, N.P.; Liu, S.; Hansen, M.; Toure, A.
2012-01-01
Spatially explicit land cover land use (LCLU) change information is needed to drive biogeochemical models that simulate soil organic carbon (SOC) dynamics. Such information is increasingly being mapped using remotely sensed satellite data with classification schemes and uncertainties constrained by the sensing system, classification algorithms and land cover schemes. In this study, automated LCLU classification of multi-temporal Landsat satellite data were used to assess the sensitivity of SOC modeled by the Global Ensemble Biogeochemical Modeling System (GEMS). The GEMS was run for an area of 1560 km2 in Senegal under three climate change scenarios with LCLU maps generated using different Landsat classification approaches. This research provides a method to estimate the variability of SOC, specifically the SOC uncertainty due to satellite classification errors, which we show is dependent not only on the LCLU classification errors but also on where the LCLU classes occur relative to the other GEMS model inputs.
Electroencephalography epilepsy classifications using hybrid cuckoo search and neural network
Pratiwi, A. B.; Damayanti, A.; Miswanto
2017-07-01
Epilepsy is a condition that affects the brain and causes repeated seizures. This seizure is episodes that can vary and nearly undetectable to long periods of vigorous shaking or brain contractions. Epilepsy often can be confirmed with an electrocephalography (EEG). Neural Networks has been used in biomedic signal analysis, it has successfully classified the biomedic signal, such as EEG signal. In this paper, a hybrid cuckoo search and neural network are used to recognize EEG signal for epilepsy classifications. The weight of the multilayer perceptron is optimized by the cuckoo search algorithm based on its error. The aim of this methods is making the network faster to obtained the local or global optimal then the process of classification become more accurate. Based on the comparison results with the traditional multilayer perceptron, the hybrid cuckoo search and multilayer perceptron provides better performance in term of error convergence and accuracy. The purpose methods give MSE 0.001 and accuracy 90.0 %.
Optimized universal color palette design for error diffusion
Kolpatzik, Bernd W.; Bouman, Charles A.
1995-04-01
Currently, many low-cost computers can only simultaneously display a palette of 256 color. However, this palette is usually selectable from a very large gamut of available colors. For many applications, this limited palette size imposes a significant constraint on the achievable image quality. We propose a method for designing an optimized universal color palette for use with halftoning methods such as error diffusion. The advantage of a universal color palette is that it is fixed and therefore allows multiple images to be displayed simultaneously. To design the palette, we employ a new vector quantization method known as sequential scalar quantization (SSQ) to allocate the colors in a visually uniform color space. The SSQ method achieves near-optimal allocation, but may be efficiently implemented using a series of lookup tables. When used with error diffusion, SSQ adds little computational overhead and may be used to minimize the visual error in an opponent color coordinate system. We compare the performance of the optimized algorithm to standard error diffusion by evaluating a visually weighted mean-squared-error measure. Our metric is based on the color difference in CIE L*AL*B*, but also accounts for the lowpass characteristic of human contrast sensitivity.
Detection and Classification of Whale Acoustic Signals
Xian, Yin
This dissertation focuses on two vital challenges in relation to whale acoustic signals: detection and classification. In detection, we evaluated the influence of the uncertain ocean environment on the spectrogram-based detector, and derived the likelihood ratio of the proposed Short Time Fourier Transform detector. Experimental results showed that the proposed detector outperforms detectors based on the spectrogram. The proposed detector is more sensitive to environmental changes because it includes phase information. In classification, our focus is on finding a robust and sparse representation of whale vocalizations. Because whale vocalizations can be modeled as polynomial phase signals, we can represent the whale calls by their polynomial phase coefficients. In this dissertation, we used the Weyl transform to capture chirp rate information, and used a two dimensional feature set to represent whale vocalizations globally. Experimental results showed that our Weyl feature set outperforms chirplet coefficients and MFCC (Mel Frequency Cepstral Coefficients) when applied to our collected data. Since whale vocalizations can be represented by polynomial phase coefficients, it is plausible that the signals lie on a manifold parameterized by these coefficients. We also studied the intrinsic structure of high dimensional whale data by exploiting its geometry. Experimental results showed that nonlinear mappings such as Laplacian Eigenmap and ISOMAP outperform linear mappings such as PCA and MDS, suggesting that the whale acoustic data is nonlinear. We also explored deep learning algorithms on whale acoustic data. We built each layer as convolutions with either a PCA filter bank (PCANet) or a DCT filter bank (DCTNet). With the DCT filter bank, each layer has different a time-frequency scale representation, and from this, one can extract different physical information. Experimental results showed that our PCANet and DCTNet achieve high classification rate on the whale
IMPACTS OF PATCH SIZE AND LANDSCAPE HETEROGENEITY ON THEMATIC IMAGE CLASSIFICATION ACCURACY
Impacts of Patch Size and Landscape Heterogeneity on Thematic Image Classification Accuracy. Currently, most thematic accuracy assessments of classified remotely sensed images oily account for errors between the various classes employed, at particular pixels of interest, thu...
Wind power error estimation in resource assessments.
Directory of Open Access Journals (Sweden)
Osvaldo Rodríguez
Full Text Available Estimating the power output is one of the elements that determine the techno-economic feasibility of a renewable project. At present, there is a need to develop reliable methods that achieve this goal, thereby contributing to wind power penetration. In this study, we propose a method for wind power error estimation based on the wind speed measurement error, probability density function, and wind turbine power curves. This method uses the actual wind speed data without prior statistical treatment based on 28 wind turbine power curves, which were fitted by Lagrange's method, to calculate the estimate wind power output and the corresponding error propagation. We found that wind speed percentage errors of 10% were propagated into the power output estimates, thereby yielding an error of 5%. The proposed error propagation complements the traditional power resource assessments. The wind power estimation error also allows us to estimate intervals for the power production leveled cost or the investment time return. The implementation of this method increases the reliability of techno-economic resource assessment studies.
Wind power error estimation in resource assessments.
Rodríguez, Osvaldo; Del Río, Jesús A; Jaramillo, Oscar A; Martínez, Manuel
2015-01-01
Estimating the power output is one of the elements that determine the techno-economic feasibility of a renewable project. At present, there is a need to develop reliable methods that achieve this goal, thereby contributing to wind power penetration. In this study, we propose a method for wind power error estimation based on the wind speed measurement error, probability density function, and wind turbine power curves. This method uses the actual wind speed data without prior statistical treatment based on 28 wind turbine power curves, which were fitted by Lagrange's method, to calculate the estimate wind power output and the corresponding error propagation. We found that wind speed percentage errors of 10% were propagated into the power output estimates, thereby yielding an error of 5%. The proposed error propagation complements the traditional power resource assessments. The wind power estimation error also allows us to estimate intervals for the power production leveled cost or the investment time return. The implementation of this method increases the reliability of techno-economic resource assessment studies.
Semiparametric Bernstein–von Mises for the error standard deviation
Jonge, de, R.; Zanten, van, J.H.
2013-01-01
We study Bayes procedures for nonparametric regression problems with Gaussian errors, giving conditions under which a Bernstein–von Mises result holds for the marginal posterior distribution of the error standard deviation. We apply our general results to show that a single Bayes procedure using a hierarchical spline-based prior on the regression function and an independent prior on the error variance, can simultaneously achieve adaptive, rate-optimal estimation of a smooth, multivariate regr...
Rokni Deilmai, B.; Ahmad, B. Bin; Zabihi, H.
2014-06-01
Mapping is essential for the analysis of the land use and land cover, which influence many environmental processes and properties. For the purpose of the creation of land cover maps, it is important to minimize error. These errors will propagate into later analyses based on these land cover maps. The reliability of land cover maps derived from remotely sensed data depends on an accurate classification. In this study, we have analyzed multispectral data using two different classifiers including Maximum Likelihood Classifier (MLC) and Support Vector Machine (SVM). To pursue this aim, Landsat Thematic Mapper data and identical field-based training sample datasets in Johor Malaysia used for each classification method, which results indicate in five land cover classes forest, oil palm, urban area, water, rubber. Classification results indicate that SVM was more accurate than MLC. With demonstrated capability to produce reliable cover results, the SVM methods should be especially useful for land cover classification.
International Nuclear Information System (INIS)
Deilmai, B Rokni; Ahmad, B Bin; Zabihi, H
2014-01-01
Mapping is essential for the analysis of the land use and land cover, which influence many environmental processes and properties. For the purpose of the creation of land cover maps, it is important to minimize error. These errors will propagate into later analyses based on these land cover maps. The reliability of land cover maps derived from remotely sensed data depends on an accurate classification. In this study, we have analyzed multispectral data using two different classifiers including Maximum Likelihood Classifier (MLC) and Support Vector Machine (SVM). To pursue this aim, Landsat Thematic Mapper data and identical field-based training sample datasets in Johor Malaysia used for each classification method, which results indicate in five land cover classes forest, oil palm, urban area, water, rubber. Classification results indicate that SVM was more accurate than MLC. With demonstrated capability to produce reliable cover results, the SVM methods should be especially useful for land cover classification
National Research Council Canada - National Science Library
Wood, Brian
2000-01-01
.... The Human Factors Analysis and Classification System-Maintenance Extension taxonomy, an effective framework for classifying and analyzing the presence of maintenance errors that lead to mishaps...
Efficient detection of dangling pointer error for C/C++ programs
Zhang, Wenzhe
2017-08-01
Dangling pointer error is pervasive in C/C++ programs and it is very hard to detect. This paper introduces an efficient detector to detect dangling pointer error in C/C++ programs. By selectively leave some memory accesses unmonitored, our method could reduce the memory monitoring overhead and thus achieves better performance over previous methods. Experiments show that our method could achieve an average speed up of 9% over previous compiler instrumentation based method and more than 50% over previous page protection based method.
Classification across gene expression microarray studies
Directory of Open Access Journals (Sweden)
Kuner Ruprecht
2009-12-01
Full Text Available Abstract Background The increasing number of gene expression microarray studies represents an important resource in biomedical research. As a result, gene expression based diagnosis has entered clinical practice for patient stratification in breast cancer. However, the integration and combined analysis of microarray studies remains still a challenge. We assessed the potential benefit of data integration on the classification accuracy and systematically evaluated the generalization performance of selected methods on four breast cancer studies comprising almost 1000 independent samples. To this end, we introduced an evaluation framework which aims to establish good statistical practice and a graphical way to monitor differences. The classification goal was to correctly predict estrogen receptor status (negative/positive and histological grade (low/high of each tumor sample in an independent study which was not used for the training. For the classification we chose support vector machines (SVM, predictive analysis of microarrays (PAM, random forest (RF and k-top scoring pairs (kTSP. Guided by considerations relevant for classification across studies we developed a generalization of kTSP which we evaluated in addition. Our derived version (DV aims to improve the robustness of the intrinsic invariance of kTSP with respect to technologies and preprocessing. Results For each individual study the generalization error was benchmarked via complete cross-validation and was found to be similar for all classification methods. The misclassification rates were substantially higher in classification across studies, when each single study was used as an independent test set while all remaining studies were combined for the training of the classifier. However, with increasing number of independent microarray studies used in the training, the overall classification performance improved. DV performed better than the average and showed slightly less variance. In
Classification of ASKAP Vast Radio Light Curves
Rebbapragada, Umaa; Lo, Kitty; Wagstaff, Kiri L.; Reed, Colorado; Murphy, Tara; Thompson, David R.
2012-01-01
The VAST survey is a wide-field survey that observes with unprecedented instrument sensitivity (0.5 mJy or lower) and repeat cadence (a goal of 5 seconds) that will enable novel scientific discoveries related to known and unknown classes of radio transients and variables. Given the unprecedented observing characteristics of VAST, it is important to estimate source classification performance, and determine best practices prior to the launch of ASKAP's BETA in 2012. The goal of this study is to identify light curve characterization and classification algorithms that are best suited for archival VAST light curve classification. We perform our experiments on light curve simulations of eight source types and achieve best case performance of approximately 90% accuracy. We note that classification performance is most influenced by light curve characterization rather than classifier algorithm.
Error analysis of the freshmen Criminology students’ grammar in the written English
Directory of Open Access Journals (Sweden)
Maico Demi Banate Aperocho
2017-12-01
Full Text Available This study identifies the various syntactical errors of the fifty (50 freshmen B.S. Criminology students of the University of Mindanao in Davao City. Specifically, this study aims to answer the following: (1 What are the common errors present in the argumentative essays of the respondents? (2 What are the reasons of the existence of these errors? This study is descriptive-qualitative. It also uses error analysis to point out the syntactical errors present in the compositions of the participants. The fifty essays are subjected to error analysis. Errors are classified based on Chanquoy’s Classification of Writing Errors. Furthermore, Hourani’s Common Reasons of Grammatical Errors Checklist was also used to determine the common reasons of the identified syntactical errors. To create a meaningful interpretation of data and to solicit further ideas from the participants, a focus group discussion is also done. Findings show that students’ most common errors are on the grammatical aspect. In the grammatical aspect, students have more frequently committed errors in the verb aspect (tense, subject agreement, and auxiliary and linker choice compared to spelling and punctuation aspects. Moreover, there are three topmost reasons of committing errors in the paragraph: mother tongue interference, incomprehensibility of the grammar rules, and the incomprehensibility of the writing mechanics. Despite the difficulty in learning English as a second language, students are still very motivated to master the concepts and applications of the language.
Texture operator for snow particle classification into snowflake and graupel
Nurzyńska, Karolina; Kubo, Mamoru; Muramoto, Ken-ichiro
2012-11-01
In order to improve the estimation of precipitation, the coefficients of Z-R relation should be determined for each snow type. Therefore, it is necessary to identify the type of falling snow. Consequently, this research addresses a problem of snow particle classification into snowflake and graupel in an automatic manner (as these types are the most common in the study region). Having correctly classified precipitation events, it is believed that it will be possible to estimate the related parameters accurately. The automatic classification system presented here describes the images with texture operators. Some of them are well-known from the literature: first order features, co-occurrence matrix, grey-tone difference matrix, run length matrix, and local binary pattern, but also a novel approach to design simple local statistic operators is introduced. In this work the following texture operators are defined: mean histogram, min-max histogram, and mean-variance histogram. Moreover, building a feature vector, which is based on the structure created in many from mentioned algorithms is also suggested. For classification, the k-nearest neighbourhood classifier was applied. The results showed that it is possible to achieve correct classification accuracy above 80% by most of the techniques. The best result of 86.06%, was achieved for operator built from a structure achieved in the middle stage of the co-occurrence matrix calculation. Next, it was noticed that describing an image with two texture operators does not improve the classification results considerably. In the best case the correct classification efficiency was 87.89% for a pair of texture operators created from local binary pattern and structure build in a middle stage of grey-tone difference matrix calculation. This also suggests that the information gathered by each texture operator is redundant. Therefore, the principal component analysis was applied in order to remove the unnecessary information and
Multiclass Classification of Cardiac Arrhythmia Using Improved Feature Selection and SVM Invariants.
Mustaqeem, Anam; Anwar, Syed Muhammad; Majid, Muahammad
2018-01-01
Arrhythmia is considered a life-threatening disease causing serious health issues in patients, when left untreated. An early diagnosis of arrhythmias would be helpful in saving lives. This study is conducted to classify patients into one of the sixteen subclasses, among which one class represents absence of disease and the other fifteen classes represent electrocardiogram records of various subtypes of arrhythmias. The research is carried out on the dataset taken from the University of California at Irvine Machine Learning Data Repository. The dataset contains a large volume of feature dimensions which are reduced using wrapper based feature selection technique. For multiclass classification, support vector machine (SVM) based approaches including one-against-one (OAO), one-against-all (OAA), and error-correction code (ECC) are employed to detect the presence and absence of arrhythmias. The SVM method results are compared with other standard machine learning classifiers using varying parameters and the performance of the classifiers is evaluated using accuracy, kappa statistics, and root mean square error. The results show that OAO method of SVM outperforms all other classifiers by achieving an accuracy rate of 81.11% when used with 80/20 data split and 92.07% using 90/10 data split option.
Analysis of error patterns in clinical radiotherapy
International Nuclear Information System (INIS)
Macklis, Roger; Meier, Tim; Barrett, Patricia; Weinhous, Martin
1996-01-01
Purpose: Until very recently, prescription errors and adverse treatment events have rarely been studied or reported systematically in oncology. We wished to understand the spectrum and severity of radiotherapy errors that take place on a day-to-day basis in a high-volume academic practice and to understand the resource needs and quality assurance challenges placed on a department by rapid upswings in contract-based clinical volumes requiring additional operating hours, procedures, and personnel. The goal was to define clinical benchmarks for operating safety and to detect error-prone treatment processes that might function as 'early warning' signs. Methods: A multi-tiered prospective and retrospective system for clinical error detection and classification was developed, with formal analysis of the antecedents and consequences of all deviations from prescribed treatment delivery, no matter how trivial. A department-wide record-and-verify system was operational during this period and was used as one method of treatment verification and error detection. Brachytherapy discrepancies were analyzed separately. Results: During the analysis year, over 2000 patients were treated with over 93,000 individual fields. A total of 59 errors affecting a total of 170 individual treated fields were reported or detected during this period. After review, all of these errors were classified as Level 1 (minor discrepancy with essentially no potential for negative clinical implications). This total treatment delivery error rate (170/93, 332 or 0.18%) is significantly better than corresponding error rates reported for other hospital and oncology treatment services, perhaps reflecting the relatively sophisticated error avoidance and detection procedures used in modern clinical radiation oncology. Error rates were independent of linac model and manufacturer, time of day (normal operating hours versus late evening or early morning) or clinical machine volumes. There was some relationship to
BANKRUPTCY PREDICTION MODEL WITH ZETAc OPTIMAL CUT-OFF SCORE TO CORRECT TYPE I ERRORS
Directory of Open Access Journals (Sweden)
Mohamad Iwan
2005-06-01
This research has successfully attained the following results: (1 type I error is in fact 59,83 times more costly compared to type II error, (2 22 ratios distinguish between bankrupt and non-bankrupt groups, (3 2 financial ratios proved to be effective in predicting bankruptcy, (4 prediction using ZETAc optimal cut-off score predicts more companies filing for bankruptcy within one year compared to prediction using Hair et al. optimum cutting score, (5 Although prediction using Hair et al. optimum cutting score is more accurate, prediction using ZETAc optimal cut-off score proved to be able to minimize cost incurred from classification errors.
Real-time, resource-constrained object classification on a micro-air vehicle
Buck, Louis; Ray, Laura
2013-12-01
A real-time embedded object classification algorithm is developed through the novel combination of binary feature descriptors, a bag-of-visual-words object model and the cortico-striatal loop (CSL) learning algorithm. The BRIEF, ORB and FREAK binary descriptors are tested and compared to SIFT descriptors with regard to their respective classification accuracies, execution times, and memory requirements when used with CSL on a 12.6 g ARM Cortex embedded processor running at 800 MHz. Additionally, the effect of x2 feature mapping and opponent-color representations used with these descriptors is examined. These tests are performed on four data sets of varying sizes and difficulty, and the BRIEF descriptor is found to yield the best combination of speed and classification accuracy. Its use with CSL achieves accuracies between 67% and 95% of those achieved with SIFT descriptors and allows for the embedded classification of a 128x192 pixel image in 0.15 seconds, 60 times faster than classification with SIFT. X2 mapping is found to provide substantial improvements in classification accuracy for all of the descriptors at little cost, while opponent-color descriptors are offer accuracy improvements only on colorful datasets.
Repeat-aware modeling and correction of short read errors.
Yang, Xiao; Aluru, Srinivas; Dorman, Karin S
2011-02-15
High-throughput short read sequencing is revolutionizing genomics and systems biology research by enabling cost-effective deep coverage sequencing of genomes and transcriptomes. Error detection and correction are crucial to many short read sequencing applications including de novo genome sequencing, genome resequencing, and digital gene expression analysis. Short read error detection is typically carried out by counting the observed frequencies of kmers in reads and validating those with frequencies exceeding a threshold. In case of genomes with high repeat content, an erroneous kmer may be frequently observed if it has few nucleotide differences with valid kmers with multiple occurrences in the genome. Error detection and correction were mostly applied to genomes with low repeat content and this remains a challenging problem for genomes with high repeat content. We develop a statistical model and a computational method for error detection and correction in the presence of genomic repeats. We propose a method to infer genomic frequencies of kmers from their observed frequencies by analyzing the misread relationships among observed kmers. We also propose a method to estimate the threshold useful for validating kmers whose estimated genomic frequency exceeds the threshold. We demonstrate that superior error detection is achieved using these methods. Furthermore, we break away from the common assumption of uniformly distributed errors within a read, and provide a framework to model position-dependent error occurrence frequencies common to many short read platforms. Lastly, we achieve better error correction in genomes with high repeat content. The software is implemented in C++ and is freely available under GNU GPL3 license and Boost Software V1.0 license at "http://aluru-sun.ece.iastate.edu/doku.php?id = redeem". We introduce a statistical framework to model sequencing errors in next-generation reads, which led to promising results in detecting and correcting errors
APPLICATION OF SENSOR FUSION TO IMPROVE UAV IMAGE CLASSIFICATION
Directory of Open Access Journals (Sweden)
S. Jabari
2017-08-01
Full Text Available Image classification is one of the most important tasks of remote sensing projects including the ones that are based on using UAV images. Improving the quality of UAV images directly affects the classification results and can save a huge amount of time and effort in this area. In this study, we show that sensor fusion can improve image quality which results in increasing the accuracy of image classification. Here, we tested two sensor fusion configurations by using a Panchromatic (Pan camera along with either a colour camera or a four-band multi-spectral (MS camera. We use the Pan camera to benefit from its higher sensitivity and the colour or MS camera to benefit from its spectral properties. The resulting images are then compared to the ones acquired by a high resolution single Bayer-pattern colour camera (here referred to as HRC. We assessed the quality of the output images by performing image classification tests. The outputs prove that the proposed sensor fusion configurations can achieve higher accuracies compared to the images of the single Bayer-pattern colour camera. Therefore, incorporating a Pan camera on-board in the UAV missions and performing image fusion can help achieving higher quality images and accordingly higher accuracy classification results.
Application of Sensor Fusion to Improve Uav Image Classification
Jabari, S.; Fathollahi, F.; Zhang, Y.
2017-08-01
Image classification is one of the most important tasks of remote sensing projects including the ones that are based on using UAV images. Improving the quality of UAV images directly affects the classification results and can save a huge amount of time and effort in this area. In this study, we show that sensor fusion can improve image quality which results in increasing the accuracy of image classification. Here, we tested two sensor fusion configurations by using a Panchromatic (Pan) camera along with either a colour camera or a four-band multi-spectral (MS) camera. We use the Pan camera to benefit from its higher sensitivity and the colour or MS camera to benefit from its spectral properties. The resulting images are then compared to the ones acquired by a high resolution single Bayer-pattern colour camera (here referred to as HRC). We assessed the quality of the output images by performing image classification tests. The outputs prove that the proposed sensor fusion configurations can achieve higher accuracies compared to the images of the single Bayer-pattern colour camera. Therefore, incorporating a Pan camera on-board in the UAV missions and performing image fusion can help achieving higher quality images and accordingly higher accuracy classification results.
Extension classification method for low-carbon product cases
Directory of Open Access Journals (Sweden)
Yanwei Zhao
2016-05-01
Full Text Available In product low-carbon design, intelligent decision systems integrated with certain classification algorithms recommend the existing design cases to designers. However, these systems mostly dependent on prior experience, and product designers not only expect to get a satisfactory case from an intelligent system but also hope to achieve assistance in modifying unsatisfactory cases. In this article, we proposed a new categorization method composed of static and dynamic classification based on extension theory. This classification method can be integrated into case-based reasoning system to get accurate classification results and to inform designers of detailed information about unsatisfactory cases. First, we establish the static classification model for cases by dependent function in a hierarchical structure. Then for dynamic classification, we make transformation for cases based on case model, attributes, attribute values, and dependent function, thus cases can take qualitative changes. Finally, the applicability of proposed method is demonstrated through a case study of screw air compressor cases.
Hu, Weiwei; Tan, Ying
2016-12-01
The nearest neighbor (NN) classifier suffers from high time complexity when classifying a test instance since the need of searching the whole training set. Prototype generation is a widely used approach to reduce the classification time, which generates a small set of prototypes to classify a test instance instead of using the whole training set. In this paper, particle swarm optimization is applied to prototype generation and two novel methods for improving the classification performance are presented: 1) a fitness function named error rank and 2) the multiobjective (MO) optimization strategy. Error rank is proposed to enhance the generation ability of the NN classifier, which takes the ranks of misclassified instances into consideration when designing the fitness function. The MO optimization strategy pursues the performance on multiple subsets of data simultaneously, in order to keep the classifier from overfitting the training set. Experimental results over 31 UCI data sets and 59 additional data sets show that the proposed algorithm outperforms nearly 30 existing prototype generation algorithms.
Subordinate-level object classification reexamined.
Biederman, I; Subramaniam, S; Bar, M; Kalocsai, P; Fiser, J
1999-01-01
The classification of a table as round rather than square, a car as a Mazda rather than a Ford, a drill bit as 3/8-inch rather than 1/4-inch, and a face as Tom have all been regarded as a single process termed "subordinate classification." Despite the common label, the considerable heterogeneity of the perceptual processing required to achieve such classifications requires, minimally, a more detailed taxonomy. Perceptual information relevant to subordinate-level shape classifications can be presumed to vary on continua of (a) the type of distinctive information that is present, nonaccidental or metric, (b) the size of the relevant contours or surfaces, and (c) the similarity of the to-be-discriminated features, such as whether a straight contour has to be distinguished from a contour of low curvature versus high curvature. We consider three, relatively pure cases. Case 1 subordinates may be distinguished by a representation, a geon structural description (GSD), specifying a nonaccidental characterization of an object's large parts and the relations among these parts, such as a round table versus a square table. Case 2 subordinates are also distinguished by GSDs, except that the distinctive GSDs are present at a small scale in a complex object so the location and mapping of the GSDs are contingent on an initial basic-level classification, such as when we use a logo to distinguish various makes of cars. Expertise for Cases 1 and 2 can be easily achieved through specification, often verbal, of the GSDs. Case 3 subordinates, which have furnished much of the grist for theorizing with "view-based" template models, require fine metric discriminations. Cases 1 and 2 account for the overwhelming majority of shape-based basic- and subordinate-level object classifications that people can and do make in their everyday lives. These classifications are typically made quickly, accurately, and with only modest costs of viewpoint changes. Whereas the activation of an array of
Human error theory: relevance to nurse management.
Armitage, Gerry
2009-03-01
Describe, discuss and critically appraise human error theory and consider its relevance for nurse managers. Healthcare errors are a persistent threat to patient safety. Effective risk management and clinical governance depends on understanding the nature of error. This paper draws upon a wide literature from published works, largely from the field of cognitive psychology and human factors. Although the content of this paper is pertinent to any healthcare professional; it is written primarily for nurse managers. Error is inevitable. Causation is often attributed to individuals, yet causation in complex environments such as healthcare is predominantly multi-factorial. Individual performance is affected by the tendency to develop prepacked solutions and attention deficits, which can in turn be related to local conditions and systems or latent failures. Blame is often inappropriate. Defences should be constructed in the light of these considerations and to promote error wisdom and organizational resilience. Managing and learning from error is seen as a priority in the British National Health Service (NHS), this can be better achieved with an understanding of the roots, nature and consequences of error. Such an understanding can provide a helpful framework for a range of risk management activities.
Research on Human-Error Factors of Civil Aircraft Pilots Based On Grey Relational Analysis
Directory of Open Access Journals (Sweden)
Guo Yundong
2018-01-01
Full Text Available In consideration of the situation that civil aviation accidents involve many human-error factors and show the features of typical grey systems, an index system of civil aviation accident human-error factors is built using human factor analysis and classification system model. With the data of accidents happened worldwide between 2008 and 2011, the correlation between human-error factors can be analyzed quantitatively using the method of grey relational analysis. Research results show that the order of main factors affecting pilot human-error factors is preconditions for unsafe acts, unsafe supervision, organization and unsafe acts. The factor related most closely with second-level indexes and pilot human-error factors is the physical/mental limitations of pilots, followed by supervisory violations. The relevancy between the first-level indexes and the corresponding second-level indexes and the relevancy between second-level indexes can also be analyzed quantitatively.
Music Genre Classification Using MIDI and Audio Features
Cataltepe, Zehra; Yaslan, Yusuf; Sonmez, Abdullah
2007-12-01
We report our findings on using MIDI files and audio features from MIDI, separately and combined together, for MIDI music genre classification. We use McKay and Fujinaga's 3-root and 9-leaf genre data set. In order to compute distances between MIDI pieces, we use normalized compression distance (NCD). NCD uses the compressed length of a string as an approximation to its Kolmogorov complexity and has previously been used for music genre and composer clustering. We convert the MIDI pieces to audio and then use the audio features to train different classifiers. MIDI and audio from MIDI classifiers alone achieve much smaller accuracies than those reported by McKay and Fujinaga who used not NCD but a number of domain-based MIDI features for their classification. Combining MIDI and audio from MIDI classifiers improves accuracy and gets closer to, but still worse, accuracies than McKay and Fujinaga's. The best root genre accuracies achieved using MIDI, audio, and combination of them are 0.75, 0.86, and 0.93, respectively, compared to 0.98 of McKay and Fujinaga. Successful classifier combination requires diversity of the base classifiers. We achieve diversity through using certain number of seconds of the MIDI file, different sample rates and sizes for the audio file, and different classification algorithms.
Managing errors in radiology: a working model
International Nuclear Information System (INIS)
Melvin, C.; Bodley, R.; Booth, A.; Meagher, T.; Record, C.; Savage, P.
2004-01-01
AIM: To develop a practical mechanism for reviewing reporting discrepancies as addressed in the Royal College of Radiologists publication 'To err is human. The case for review of reporting discrepancies'. MATERIALS AND METHODS: A regular meeting was developed, and has evolved, within the department to review discrepancies. Standard forms were devised for submission of cases as well as recording and classification of discrepancies. This has resulted in availability of figures that can be audited annually. RESULTS: Eighty-one cases involving error were reviewed over a 12-month period. Seven further cases flagged as discrepancies were not identified on peer review. Twenty-four reports were amended subsequent to the meeting. Nineteen additional cases were brought to the meeting as illustrative of teaching points or for discussion. CONCLUSION: We have evolved a successful process of reviewing reporting errors, which enjoys the confidence and support of all clinical radiologists, and is perceived as a method of improving patient care through an increasing awareness of lapses in performance
SHIP CLASSIFICATION FROM MULTISPECTRAL VIDEOS
Directory of Open Access Journals (Sweden)
Frederique Robert-Inacio
2012-05-01
Full Text Available Surveillance of a seaport can be achieved by different means: radar, sonar, cameras, radio communications and so on. Such a surveillance aims, on the one hand, to manage cargo and tanker traffic, and, on the other hand, to prevent terrorist attacks in sensitive areas. In this paper an application to video-surveillance of a seaport entrance is presented, and more particularly, the different steps enabling to classify mobile shapes. This classification is based on a parameter measuring the similarity degree between the shape under study and a set of reference shapes. The classification result describes the considered mobile in terms of shape and speed.
Habibzadeh, Mehdi; Jannesari, Mahboobeh; Rezaei, Zahra; Baharvand, Hossein; Totonchi, Mehdi
2018-04-01
This works gives an account of evaluation of white blood cell differential counts via computer aided diagnosis (CAD) system and hematology rules. Leukocytes, also called white blood cells (WBCs) play main role of the immune system. Leukocyte is responsible for phagocytosis and immunity and therefore in defense against infection involving the fatal diseases incidence and mortality related issues. Admittedly, microscopic examination of blood samples is a time consuming, expensive and error-prone task. A manual diagnosis would search for specific Leukocytes and number abnormalities in the blood slides while complete blood count (CBC) examination is performed. Complications may arise from the large number of varying samples including different types of Leukocytes, related sub-types and concentration in blood, which makes the analysis prone to human error. This process can be automated by computerized techniques which are more reliable and economical. In essence, we seek to determine a fast, accurate mechanism for classification and gather information about distribution of white blood evidences which may help to diagnose the degree of any abnormalities during CBC test. In this work, we consider the problem of pre-processing and supervised classification of white blood cells into their four primary types including Neutrophils, Eosinophils, Lymphocytes, and Monocytes using a consecutive proposed deep learning framework. For first step, this research proposes three consecutive pre-processing calculations namely are color distortion; bounding box distortion (crop) and image flipping mirroring. In second phase, white blood cell recognition performed with hierarchy topological feature extraction using Inception and ResNet architectures. Finally, the results obtained from the preliminary analysis of cell classification with (11200) training samples and 1244 white blood cells evaluation data set are presented in confusion matrices and interpreted using accuracy rate, and false
Definition and classification of epilepsy. Classification of epileptic seizures 2016
Directory of Open Access Journals (Sweden)
K. Yu. Mukhin
2017-01-01
Full Text Available Epilepsy is one of the most common neurological diseases, especially in childhood and adolescence. The incidence varies from 15 to 113 cases per 100 000 population with the maximum among children under 1 year old. The prevalence of epilepsy is high, ranging from 5 to 8 cases (in some regions – 10 cases per 1000 children under 15 years old. Classification of the disease has great importance for diagnosis, treatment and prognosis. The article presents a novel strategy for classification of epileptic seizures, developed in 2016. It contains a number of brand new concepts, including a very important one, saying that some seizures, previously considered as generalized or focal only, can be, in fact, both focal and generalized. They include tonic, atonic, myoclonic seizures and epileptic spasms. The term “secondarily generalized seizure” is replace by the term “bilateral tonic-clonic seizure” (as soon as it is not a separate type of epileptic seizures, and the term reflects the spread of discharge from any area of cerebral cortex and evolution of any types of focal seizures. International League Against Epilepsy recommends to abandon the term “pseudo-epileptic seizures” and replace it by the term “psychogenic non-epileptic seizures”. If a doctor is not sure that seizures have epileptic nature, the term “paroxysmal event” should be used without specifying the disease. The conception of childhood epileptic encephalopathies, developed within this novel classification project, is one of the most significant achievements, since in this case not only the seizures, but even epileptiform activity can induce severe disorders of higher mental functions. In addition to detailed description of the new strategy for classification of epileptic seizures, the article contains a comprehensive review of the existing principles of epilepsy and epileptic seizures classification.
Locality-preserving sparse representation-based classification in hyperspectral imagery
Gao, Lianru; Yu, Haoyang; Zhang, Bing; Li, Qingting
2016-10-01
This paper proposes to combine locality-preserving projections (LPP) and sparse representation (SR) for hyperspectral image classification. The LPP is first used to reduce the dimensionality of all the training and testing data by finding the optimal linear approximations to the eigenfunctions of the Laplace Beltrami operator on the manifold, where the high-dimensional data lies. Then, SR codes the projected testing pixels as sparse linear combinations of all the training samples to classify the testing pixels by evaluating which class leads to the minimum approximation error. The integration of LPP and SR represents an innovative contribution to the literature. The proposed approach, called locality-preserving SR-based classification, addresses the imbalance between high dimensionality of hyperspectral data and the limited number of training samples. Experimental results on three real hyperspectral data sets demonstrate that the proposed approach outperforms the original counterpart, i.e., SR-based classification.
Correa Bahnsen, Alejandro
2015-01-01
Several real-world binary classification problems are example-dependent cost-sensitive in nature, where the costs due to misclassification vary between examples and not only within classes. However, standard binary classification methods do not take these costs into account, and assume a constant cost of misclassification errors. This approach is not realistic in many real-world applications. For example in credit card fraud detection, failing to detect a fraudulent transaction may have an ec...
EEG Eye State Identification Using Incremental Attribute Learning with Time-Series Classification
Directory of Open Access Journals (Sweden)
Ting Wang
2014-01-01
Full Text Available Eye state identification is a kind of common time-series classification problem which is also a hot spot in recent research. Electroencephalography (EEG is widely used in eye state classification to detect human's cognition state. Previous research has validated the feasibility of machine learning and statistical approaches for EEG eye state classification. This paper aims to propose a novel approach for EEG eye state identification using incremental attribute learning (IAL based on neural networks. IAL is a novel machine learning strategy which gradually imports and trains features one by one. Previous studies have verified that such an approach is applicable for solving a number of pattern recognition problems. However, in these previous works, little research on IAL focused on its application to time-series problems. Therefore, it is still unknown whether IAL can be employed to cope with time-series problems like EEG eye state classification. Experimental results in this study demonstrates that, with proper feature extraction and feature ordering, IAL can not only efficiently cope with time-series classification problems, but also exhibit better classification performance in terms of classification error rates in comparison with conventional and some other approaches.
Directory of Open Access Journals (Sweden)
Alejandro Gonzalez
2014-01-01
Full Text Available Brain-machine interfaces (BMI rely on the accurate classification of event-related potentials (ERPs and their performance greatly depends on the appropriate selection of classifier parameters and features from dense-array electroencephalography (EEG signals. Moreover, in order to achieve a portable and more compact BMI for practical applications, it is also desirable to use a system capable of accurate classification using information from as few EEG channels as possible. In the present work, we propose a method for classifying P300 ERPs using a combination of Fisher Discriminant Analysis (FDA and a multiobjective hybrid real-binary Particle Swarm Optimization (MHPSO algorithm. Specifically, the algorithm searches for the set of EEG channels and classifier parameters that simultaneously maximize the classification accuracy and minimize the number of used channels. The performance of the method is assessed through offline analyses on datasets of auditory ERPs from sound discrimination experiments. The proposed method achieved a higher classification accuracy than that achieved by traditional methods while also using fewer channels. It was also found that the number of channels used for classification can be significantly reduced without greatly compromising the classification accuracy.
Kessels-Habraken, M.M.P.; Schaaf, van der T.W.; Jonge, de J.; Rutte, C.G.
2010-01-01
Medical errors in health care still occur frequently. Unfortunately, errors cannot be completely prevented and 100% safety can never be achieved. Therefore, in addition to error reduction strategies, health care organisations could also implement strategies that promote timely error detection and
LEARNING FROM MISTAKES Error Analysis in the English Speech of Indonesian Tertiary Students
Directory of Open Access Journals (Sweden)
Imelda Gozali
2017-12-01
Full Text Available This study is part of a series of Classroom Action Research conducted with the aim of improving the English speech of students in one of the tertiary institutes in Indonesia. After some years of teaching English conversation, the writer noted that students made various types of errors in their speech, which can be classified generally into morphological, phonological, and lexical. While some of the errors are still generally acceptable, some others elicit laughter or inhibit comprehension altogether. Therefore, the writer is keen to analyze the more common errors made by the students, so as to be able to compile a teaching material that could be utilized to address those errors more effectively in future classes. This research used Error Analysis by Richards (1971 as the basis of classification. It was carried out in five classes with a total number of 80 students for a period of one semester (14 weeks. The results showed that most of the errors were phonological (errors in pronunciation, while others were morphological or grammatical in nature. This prompted the writer to design simple Phonics lessons for future classes.
Rodrigo, Ma. Mercedes T.; Andallaza, Thor Collin S.; Castro, Francisco Enrique Vicente G.; Armenta, Marc Lester V.; Dy, Thomas T.; Jadud, Matthew C.
2013-01-01
In this article we quantitatively and qualitatively analyze a sample of novice programmer compilation log data, exploring whether (or how) low-achieving, average, and high-achieving students vary in their grasp of these introductory concepts. High-achieving students self-reported having the easiest time learning the introductory programming…
Directory of Open Access Journals (Sweden)
Ganchimeg Ganbold
2017-03-01
Full Text Available There are several statistical classification algorithms available for landuse/land cover classification. However, each has a certain bias orcompromise. Some methods like the parallel piped approach in supervisedclassification, cannot classify continuous regions within a feature. Onthe other hand, while unsupervised classification method takes maximumadvantage of spectral variability in an image, the maximally separableclusters in spectral space may not do much for our perception of importantclasses in a given study area. In this research, the output of an ANNalgorithm was compared with the Possibilistic c-Means an improvementof the fuzzy c-Means on both moderate resolutions Landsat8 and a highresolution Formosat 2 images. The Formosat 2 image comes with an8m spectral resolution on the multispectral data. This multispectral imagedata was resampled to 10m in order to maintain a uniform ratio of1:3 against Landsat 8 image. Six classes were chosen for analysis including:Dense forest, eucalyptus, water, grassland, wheat and riverine sand. Using a standard false color composite (FCC, the six features reflecteddifferently in the infrared region with wheat producing the brightestpixel values. Signature collection per class was therefore easily obtainedfor all classifications. The output of both ANN and FCM, were analyzedseparately for accuracy and an error matrix generated to assess the qualityand accuracy of the classification algorithms. When you compare theresults of the two methods on a per-class-basis, ANN had a crisperoutput compared to PCM which yielded clusters with pixels especiallyon the moderate resolution Landsat 8 imagery.
Multispectral LiDAR Data for Land Cover Classification of Urban Areas
Directory of Open Access Journals (Sweden)
Salem Morsy
2017-04-01
Full Text Available Airborne Light Detection And Ranging (LiDAR systems usually operate at a monochromatic wavelength measuring the range and the strength of the reflected energy (intensity from objects. Recently, multispectral LiDAR sensors, which acquire data at different wavelengths, have emerged. This allows for recording of a diversity of spectral reflectance from objects. In this context, we aim to investigate the use of multispectral LiDAR data in land cover classification using two different techniques. The first is image-based classification, where intensity and height images are created from LiDAR points and then a maximum likelihood classifier is applied. The second is point-based classification, where ground filtering and Normalized Difference Vegetation Indices (NDVIs computation are conducted. A dataset of an urban area located in Oshawa, Ontario, Canada, is classified into four classes: buildings, trees, roads and grass. An overall accuracy of up to 89.9% and 92.7% is achieved from image classification and 3D point classification, respectively. A radiometric correction model is also applied to the intensity data in order to remove the attenuation due to the system distortion and terrain height variation. The classification process is then repeated, and the results demonstrate that there are no significant improvements achieved in the overall accuracy.
Multispectral LiDAR Data for Land Cover Classification of Urban Areas.
Morsy, Salem; Shaker, Ahmed; El-Rabbany, Ahmed
2017-04-26
Airborne Light Detection And Ranging (LiDAR) systems usually operate at a monochromatic wavelength measuring the range and the strength of the reflected energy (intensity) from objects. Recently, multispectral LiDAR sensors, which acquire data at different wavelengths, have emerged. This allows for recording of a diversity of spectral reflectance from objects. In this context, we aim to investigate the use of multispectral LiDAR data in land cover classification using two different techniques. The first is image-based classification, where intensity and height images are created from LiDAR points and then a maximum likelihood classifier is applied. The second is point-based classification, where ground filtering and Normalized Difference Vegetation Indices (NDVIs) computation are conducted. A dataset of an urban area located in Oshawa, Ontario, Canada, is classified into four classes: buildings, trees, roads and grass. An overall accuracy of up to 89.9% and 92.7% is achieved from image classification and 3D point classification, respectively. A radiometric correction model is also applied to the intensity data in order to remove the attenuation due to the system distortion and terrain height variation. The classification process is then repeated, and the results demonstrate that there are no significant improvements achieved in the overall accuracy.
Lacie phase 1 Classification and Mensuration Subsystem (CAMS) rework experiment
Chhikara, R. S.; Hsu, E. M.; Liszcz, C. J.
1976-01-01
An experiment was designed to test the ability of the Classification and Mensuration Subsystem rework operations to improve wheat proportion estimates for segments that had been processed previously. Sites selected for the experiment included three in Kansas and three in Texas, with the remaining five distributed in Montana and North and South Dakota. The acquisition dates were selected to be representative of imagery available in actual operations. No more than one acquisition per biophase were used, and biophases were determined by actual crop calendars. All sites were worked by each of four Analyst-Interpreter/Data Processing Analyst Teams who reviewed the initial processing of each segment and accepted or reworked it for an estimate of the proportion of small grains in the segment. Classification results, acquisitions and classification errors and performance results between CAMS regular and ITS rework are tabulated.
Basic Hand Gestures Classification Based on Surface Electromyography
Directory of Open Access Journals (Sweden)
Aleksander Palkowski
2016-01-01
Full Text Available This paper presents an innovative classification system for hand gestures using 2-channel surface electromyography analysis. The system developed uses the Support Vector Machine classifier, for which the kernel function and parameter optimisation are conducted additionally by the Cuckoo Search swarm algorithm. The system developed is compared with standard Support Vector Machine classifiers with various kernel functions. The average classification rate of 98.12% has been achieved for the proposed method.
Medical errors in hospitalized pediatric trauma patients with chronic health conditions
Directory of Open Access Journals (Sweden)
Xiaotong Liu
2014-01-01
Full Text Available Objective: This study compares medical errors in pediatric trauma patients with and without chronic conditions. Methods: The 2009 Kids’ Inpatient Database, which included 123,303 trauma discharges, was analyzed. Medical errors were identified by International Classification of Diseases, Ninth Revision, Clinical Modification diagnosis codes. The medical error rates per 100 discharges and per 1000 hospital days were calculated and compared between inpatients with and without chronic conditions. Results: Pediatric trauma patients with chronic conditions experienced a higher medical error rate compared with patients without chronic conditions: 4.04 (95% confidence interval: 3.75–4.33 versus 1.07 (95% confidence interval: 0.98–1.16 per 100 discharges. The rate of medical error differed by type of chronic condition. After controlling for confounding factors, the presence of a chronic condition increased the adjusted odds ratio of medical error by 37% if one chronic condition existed (adjusted odds ratio: 1.37, 95% confidence interval: 1.21–1.5, and 69% if more than one chronic condition existed (adjusted odds ratio: 1.69, 95% confidence interval: 1.48–1.53. In the adjusted model, length of stay had the strongest association with medical error, but the adjusted odds ratio for chronic conditions and medical error remained significantly elevated even when accounting for the length of stay, suggesting that medical complexity has a role in medical error. Higher adjusted odds ratios were seen in other subgroups. Conclusion: Chronic conditions are associated with significantly higher rate of medical errors in pediatric trauma patients. Future research should evaluate interventions or guidelines for reducing the risk of medical errors in pediatric trauma patients with chronic conditions.
The nearest neighbor and the bayes error rates.
Loizou, G; Maybank, S J
1987-02-01
The (k, l) nearest neighbor method of pattern classification is compared to the Bayes method. If the two acceptance rates are equal then the asymptotic error rates satisfy the inequalities Ek,l + 1 ¿ E*(¿) ¿ Ek,l dE*(¿), where d is a function of k, l, and the number of pattern classes, and ¿ is the reject threshold for the Bayes method. An explicit expression for d is given which is optimal in the sense that for some probability distributions Ek,l and dE* (¿) are equal.
Fernández-Delgado, Manuel; Cernadas, Eva; Barro, Senén; Ribeiro, Jorge; Neves, José
2014-02-01
The Direct Kernel Perceptron (DKP) (Fernández-Delgado et al., 2010) is a very simple and fast kernel-based classifier, related to the Support Vector Machine (SVM) and to the Extreme Learning Machine (ELM) (Huang, Wang, & Lan, 2011), whose α-coefficients are calculated directly, without any iterative training, using an analytical closed-form expression which involves only the training patterns. The DKP, which is inspired by the Direct Parallel Perceptron, (Auer et al., 2008), uses a Gaussian kernel and a linear classifier (perceptron). The weight vector of this classifier in the feature space minimizes an error measure which combines the training error and the hyperplane margin, without any tunable regularization parameter. This weight vector can be translated, using a variable change, to the α-coefficients, and both are determined without iterative calculations. We calculate solutions using several error functions, achieving the best trade-off between accuracy and efficiency with the linear function. These solutions for the α coefficients can be considered alternatives to the ELM with a new physical meaning in terms of error and margin: in fact, the linear and quadratic DKP are special cases of the two-class ELM when the regularization parameter C takes the values C=0 and C=∞. The linear DKP is extremely efficient and much faster (over a vast collection of 42 benchmark and real-life data sets) than 12 very popular and accurate classifiers including SVM, Multi-Layer Perceptron, Adaboost, Random Forest and Bagging of RPART decision trees, Linear Discriminant Analysis, K-Nearest Neighbors, ELM, Probabilistic Neural Networks, Radial Basis Function neural networks and Generalized ART. Besides, despite its simplicity and extreme efficiency, DKP achieves higher accuracies than 7 out of 12 classifiers, exhibiting small differences with respect to the best ones (SVM, ELM, Adaboost and Random Forest), which are much slower. Thus, the DKP provides an easy and fast way
Directory of Open Access Journals (Sweden)
Francisco J Valverde-Albacete
Full Text Available The most widely spread measure of performance, accuracy, suffers from a paradox: predictive models with a given level of accuracy may have greater predictive power than models with higher accuracy. Despite optimizing classification error rate, high accuracy models may fail to capture crucial information transfer in the classification task. We present evidence of this behavior by means of a combinatorial analysis where every possible contingency matrix of 2, 3 and 4 classes classifiers are depicted on the entropy triangle, a more reliable information-theoretic tool for classification assessment. Motivated by this, we develop from first principles a measure of classification performance that takes into consideration the information learned by classifiers. We are then able to obtain the entropy-modulated accuracy (EMA, a pessimistic estimate of the expected accuracy with the influence of the input distribution factored out, and the normalized information transfer factor (NIT, a measure of how efficient is the transmission of information from the input to the output set of classes. The EMA is a more natural measure of classification performance than accuracy when the heuristic to maximize is the transfer of information through the classifier instead of classification error count. The NIT factor measures the effectiveness of the learning process in classifiers and also makes it harder for them to "cheat" using techniques like specialization, while also promoting the interpretability of results. Their use is demonstrated in a mind reading task competition that aims at decoding the identity of a video stimulus based on magnetoencephalography recordings. We show how the EMA and the NIT factor reject rankings based in accuracy, choosing more meaningful and interpretable classifiers.
Completed Local Ternary Pattern for Rotation Invariant Texture Classification
Directory of Open Access Journals (Sweden)
Taha H. Rassem
2014-01-01
Full Text Available Despite the fact that the two texture descriptors, the completed modeling of Local Binary Pattern (CLBP and the Completed Local Binary Count (CLBC, have achieved a remarkable accuracy for invariant rotation texture classification, they inherit some Local Binary Pattern (LBP drawbacks. The LBP is sensitive to noise, and different patterns of LBP may be classified into the same class that reduces its discriminating property. Although, the Local Ternary Pattern (LTP is proposed to be more robust to noise than LBP, however, the latter’s weakness may appear with the LTP as well as with LBP. In this paper, a novel completed modeling of the Local Ternary Pattern (LTP operator is proposed to overcome both LBP drawbacks, and an associated completed Local Ternary Pattern (CLTP scheme is developed for rotation invariant texture classification. The experimental results using four different texture databases show that the proposed CLTP achieved an impressive classification accuracy as compared to the CLBP and CLBC descriptors.
National Research Council Canada - National Science Library
Boex, Anthony
2001-01-01
.... The Human Factors Analysis and Classification System-Maintenance Extension (HFACS-ME) taxonomy, a framework for classifying and analyzing the presence of maintenance errors that lead to mishaps, is the foundation of this tool...
Research and practice on NPP safety DCS application software V and V defect classification system
International Nuclear Information System (INIS)
Zhang Dongwei; Li Yunjian; Li Xiangjian
2012-01-01
One of the most significant aims of Verification and Validation (V and V) is to find software errors and risks, especially for a DCS application software designed for nuclear power plant (NPP). Through classifying and analyzing errors, a number of obtained data can be utilized to estimate current status and potential risks of software development and improve the quality of project. A method of error classification is proposed, which is applied to whole V and V life cycle, using a MW pressurized reactor project as an example. The purpose is to analyze errors discovered by V and V activities, and result in improvement of safety critical DCS application software. (authors)
Error modeling for surrogates of dynamical systems using machine learning
Trehan, Sumeet; Carlberg, Kevin T.; Durlofsky, Louis J.
2017-12-01
A machine-learning-based framework for modeling the error introduced by surrogate models of parameterized dynamical systems is proposed. The framework entails the use of high-dimensional regression techniques (e.g., random forests, LASSO) to map a large set of inexpensively computed `error indicators' (i.e., features) produced by the surrogate model at a given time instance to a prediction of the surrogate-model error in a quantity of interest (QoI). This eliminates the need for the user to hand-select a small number of informative features. The methodology requires a training set of parameter instances at which the time-dependent surrogate-model error is computed by simulating both the high-fidelity and surrogate models. Using these training data, the method first determines regression-model locality (via classification or clustering), and subsequently constructs a `local' regression model to predict the time-instantaneous error within each identified region of feature space. We consider two uses for the resulting error model: (1) as a correction to the surrogate-model QoI prediction at each time instance, and (2) as a way to statistically model arbitrary functions of the time-dependent surrogate-model error (e.g., time-integrated errors). We apply the proposed framework to model errors in reduced-order models of nonlinear oil--water subsurface flow simulations. The reduced-order models used in this work entail application of trajectory piecewise linearization with proper orthogonal decomposition. When the first use of the method is considered, numerical experiments demonstrate consistent improvement in accuracy in the time-instantaneous QoI prediction relative to the original surrogate model, across a large number of test cases. When the second use is considered, results show that the proposed method provides accurate statistical predictions of the time- and well-averaged errors.
Alheadary, Wael Ghazy
2016-12-24
In this work, we present a bit error rate (BER) and achievable spectral efficiency (ASE) performance of a freespace optical (FSO) link with pointing errors based on intensity modulation/direct detection (IM/DD) and heterodyne detection over general Malaga turbulence channel. More specifically, we present exact closed-form expressions for adaptive and non-adaptive transmission. The closed form expressions are presented in terms of generalized power series of the Meijer\\'s G-function. Moreover, asymptotic closed form expressions are provided to validate our work. In addition, all the presented analytical results are illustrated using a selected set of numerical results.
Subclinical naming errors in mild cognitive impairment: A semantic deficit?
Directory of Open Access Journals (Sweden)
Indra F. Willers
Full Text Available Abstract Mild cognitive impairment (MCI is the transitional stage between normal aging and Alzheimer's disease (AD. Impairments in semantic memory have been demonstrated to be a critical factor in early AD. The Boston Naming Test (BNT is a straightforward method of examining semantic or visuo-perceptual processing and therefore represents a potential diagnostic tool. The objective of this study was to examine naming ability and identify error types in patients with amnestic mild cognitive impairment (aMCI. Methods: Twenty aMCI patients, twenty AD patients and twenty-one normal controls, matched by age, sex and education level were evaluated. As part of a further neuropsychological evaluation, all subjects performed the BNT. A comprehensive classification of error types was devised in order to compare performance and ascertain semantic or perceptual origin of errors. Results: AD patients obtained significantly lower total scores on the BNT than aMCI patients and controls. aMCI patients did not obtain significant differences in total scores, but showed significantly higher semantic errors compared to controls. Conclusion: This study reveals that semantic processing is impaired during confrontation naming in aMCI.
International Nuclear Information System (INIS)
Yun, Hae-Bum; Masri, Sami F
2009-01-01
A reliable structural health monitoring methodology (SHM) is proposed to detect relatively small changes in uncertain nonlinear systems. A total of 4000 physical tests were performed using a complex nonlinear magneto-rheological (MR) damper. With the effective (or 'genuine') changes and uncertainties in the system characteristics of the semi-active MR damper, which were precisely controlled with known means and standard deviation of the input current, the tested MR damper was identified with the restoring force method (RFM), a non-parametric system identification method involving two-dimensional orthogonal polynomials. Using the identified RFM coefficients, both supervised and unsupervised pattern recognition techniques (including support vector classification and k-means clustering) were employed to detect system changes in the MR damper. The classification results showed that the identified coefficients with orthogonal basis function can be used as reliable indicators for detecting (small) changes, interpreting the physical meaning of the detected changes without a priori knowledge of the monitored system and quantifying the uncertainty bounds of the detected changes. The classification errors were analyzed using the standard detection theory to evaluate the performance of the developed SHM methodology. An optimal classifier design procedure was also proposed and evaluated to minimize type II (or 'missed') errors
The computation of equating errors in international surveys in education.
Monseur, Christian; Berezner, Alla
2007-01-01
Since the IEA's Third International Mathematics and Science Study, one of the major objectives of international surveys in education has been to report trends in achievement. The names of the two current IEA surveys reflect this growing interest: Trends in International Mathematics and Science Study (TIMSS) and Progress in International Reading Literacy Study (PIRLS). Similarly a central concern of the OECD's PISA is with trends in outcomes over time. To facilitate trend analyses these studies link their tests using common item equating in conjunction with item response modelling methods. IEA and PISA policies differ in terms of reporting the error associated with trends. In IEA surveys, the standard errors of the trend estimates do not include the uncertainty associated with the linking step while PISA does include a linking error component in the standard errors of trend estimates. In other words, PISA implicitly acknowledges that trend estimates partly depend on the selected common items, while the IEA's surveys do not recognise this source of error. Failing to recognise the linking error leads to an underestimation of the standard errors and thus increases the Type I error rate, thereby resulting in reporting of significant changes in achievement when in fact these are not significant. The growing interest of policy makers in trend indicators and the impact of the evaluation of educational reforms appear to be incompatible with such underestimation. However, the procedure implemented by PISA raises a few issues about the underlying assumptions for the computation of the equating error. After a brief introduction, this paper will describe the procedure PISA implemented to compute the linking error. The underlying assumptions of this procedure will then be discussed. Finally an alternative method based on replication techniques will be presented, based on a simulation study and then applied to the PISA 2000 data.
Integrating Globality and Locality for Robust Representation Based Classification
Directory of Open Access Journals (Sweden)
Zheng Zhang
2014-01-01
Full Text Available The representation based classification method (RBCM has shown huge potential for face recognition since it first emerged. Linear regression classification (LRC method and collaborative representation classification (CRC method are two well-known RBCMs. LRC and CRC exploit training samples of each class and all the training samples to represent the testing sample, respectively, and subsequently conduct classification on the basis of the representation residual. LRC method can be viewed as a “locality representation” method because it just uses the training samples of each class to represent the testing sample and it cannot embody the effectiveness of the “globality representation.” On the contrary, it seems that CRC method cannot own the benefit of locality of the general RBCM. Thus we propose to integrate CRC and LRC to perform more robust representation based classification. The experimental results on benchmark face databases substantially demonstrate that the proposed method achieves high classification accuracy.
Remote Sensing Image Classification Based on Stacked Denoising Autoencoder
Directory of Open Access Journals (Sweden)
Peng Liang
2017-12-01
Full Text Available Focused on the issue that conventional remote sensing image classification methods have run into the bottlenecks in accuracy, a new remote sensing image classification method inspired by deep learning is proposed, which is based on Stacked Denoising Autoencoder. First, the deep network model is built through the stacked layers of Denoising Autoencoder. Then, with noised input, the unsupervised Greedy layer-wise training algorithm is used to train each layer in turn for more robust expressing, characteristics are obtained in supervised learning by Back Propagation (BP neural network, and the whole network is optimized by error back propagation. Finally, Gaofen-1 satellite (GF-1 remote sensing data are used for evaluation, and the total accuracy and kappa accuracy reach 95.7% and 0.955, respectively, which are higher than that of the Support Vector Machine and Back Propagation neural network. The experiment results show that the proposed method can effectively improve the accuracy of remote sensing image classification.
Proportionate Minimum Error Entropy Algorithm for Sparse System Identification
Directory of Open Access Journals (Sweden)
Zongze Wu
2015-08-01
Full Text Available Sparse system identification has received a great deal of attention due to its broad applicability. The proportionate normalized least mean square (PNLMS algorithm, as a popular tool, achieves excellent performance for sparse system identification. In previous studies, most of the cost functions used in proportionate-type sparse adaptive algorithms are based on the mean square error (MSE criterion, which is optimal only when the measurement noise is Gaussian. However, this condition does not hold in most real-world environments. In this work, we use the minimum error entropy (MEE criterion, an alternative to the conventional MSE criterion, to develop the proportionate minimum error entropy (PMEE algorithm for sparse system identification, which may achieve much better performance than the MSE based methods especially in heavy-tailed non-Gaussian situations. Moreover, we analyze the convergence of the proposed algorithm and derive a sufficient condition that ensures the mean square convergence. Simulation results confirm the excellent performance of the new algorithm.
International Nuclear Information System (INIS)
Johnson, Sarah J; Ong, Lawrence; Shirvanimoghaddam, Mahyar; Lance, Andrew M; Symul, Thomas; Ralph, T C
2017-01-01
The maximum operational range of continuous variable quantum key distribution protocols has shown to be improved by employing high-efficiency forward error correction codes. Typically, the secret key rate model for such protocols is modified to account for the non-zero word error rate of such codes. In this paper, we demonstrate that this model is incorrect: firstly, we show by example that fixed-rate error correction codes, as currently defined, can exhibit efficiencies greater than unity. Secondly, we show that using this secret key model combined with greater than unity efficiency codes, implies that it is possible to achieve a positive secret key over an entanglement breaking channel—an impossible scenario. We then consider the secret key model from a post-selection perspective, and examine the implications for key rate if we constrain the forward error correction codes to operate at low word error rates. (paper)
Key-phrase based classification of public health web pages.
Dolamic, Ljiljana; Boyer, Célia
2013-01-01
This paper describes and evaluates the public health web pages classification model based on key phrase extraction and matching. Easily extendible both in terms of new classes as well as the new language this method proves to be a good solution for text classification faced with the total lack of training data. To evaluate the proposed solution we have used a small collection of public health related web pages created by a double blind manual classification. Our experiments have shown that by choosing the adequate threshold value the desired value for either precision or recall can be achieved.
Music Genre Classification Using MIDI and Audio Features
Directory of Open Access Journals (Sweden)
Abdullah Sonmez
2007-01-01
Full Text Available We report our findings on using MIDI files and audio features from MIDI, separately and combined together, for MIDI music genre classification. We use McKay and Fujinaga's 3-root and 9-leaf genre data set. In order to compute distances between MIDI pieces, we use normalized compression distance (NCD. NCD uses the compressed length of a string as an approximation to its Kolmogorov complexity and has previously been used for music genre and composer clustering. We convert the MIDI pieces to audio and then use the audio features to train different classifiers. MIDI and audio from MIDI classifiers alone achieve much smaller accuracies than those reported by McKay and Fujinaga who used not NCD but a number of domain-based MIDI features for their classification. Combining MIDI and audio from MIDI classifiers improves accuracy and gets closer to, but still worse, accuracies than McKay and Fujinaga's. The best root genre accuracies achieved using MIDI, audio, and combination of them are 0.75, 0.86, and 0.93, respectively, compared to 0.98 of McKay and Fujinaga. Successful classifier combination requires diversity of the base classifiers. We achieve diversity through using certain number of seconds of the MIDI file, different sample rates and sizes for the audio file, and different classification algorithms.
Action errors, error management, and learning in organizations.
Frese, Michael; Keith, Nina
2015-01-03
Every organization is confronted with errors. Most errors are corrected easily, but some may lead to negative consequences. Organizations often focus on error prevention as a single strategy for dealing with errors. Our review suggests that error prevention needs to be supplemented by error management--an approach directed at effectively dealing with errors after they have occurred, with the goal of minimizing negative and maximizing positive error consequences (examples of the latter are learning and innovations). After defining errors and related concepts, we review research on error-related processes affected by error management (error detection, damage control). Empirical evidence on positive effects of error management in individuals and organizations is then discussed, along with emotional, motivational, cognitive, and behavioral pathways of these effects. Learning from errors is central, but like other positive consequences, learning occurs under certain circumstances--one being the development of a mind-set of acceptance of human error.
Free classification of regional dialects of American English
Clopper, Cynthia G.; Pisoni, David B.
2011-01-01
Recent studies have found that naïve listeners perform poorly in forced-choice dialect categorization tasks. However, the listeners' error patterns in these tasks reveal systematic confusions between phonologically similar dialects. In the present study, a free classification procedure was used to measure the perceptual similarity structure of regional dialect variation in the United States. In two experiments, participants listened to a set of short English sentences produced by male talkers only (Experiment 1) and by male and female talkers (Experiment 2). The listeners were instructed to group the talkers by regional dialect into as many groups as they wanted with as many talkers in each group as they wished. Multidimensional scaling analyses of the data revealed three primary dimensions of perceptual similarity (linguistic markedness, geography, and gender). In addition, a comparison of the results obtained from the free classification task to previous results using the same stimulus materials in six-alternative forced-choice categorization tasks revealed that response biases in the six-alternative task were reduced or eliminated in the free classification task. Thus, the results obtained with the free classification task in the current study provided further evidence that the underlying structure of perceptual dialect category representations reflects important linguistic and sociolinguistic factors. PMID:21423862
Free classification of regional dialects of American English.
Clopper, Cynthia G; Pisoni, David B
2007-07-01
Recent studies have found that naïve listeners perform poorly in forced-choice dialect categorization tasks. However, the listeners' error patterns in these tasks reveal systematic confusions between phonologically similar dialects. In the present study, a free classification procedure was used to measure the perceptual similarity structure of regional dialect variation in the United States. In two experiments, participants listened to a set of short English sentences produced by male talkers only (Experiment 1) and by male and female talkers (Experiment 2). The listeners were instructed to group the talkers by regional dialect into as many groups as they wanted with as many talkers in each group as they wished. Multidimensional scaling analyses of the data revealed three primary dimensions of perceptual similarity (linguistic markedness, geography, and gender). In addition, a comparison of the results obtained from the free classification task to previous results using the same stimulus materials in six-alternative forced-choice categorization tasks revealed that response biases in the six-alternative task were reduced or eliminated in the free classification task. Thus, the results obtained with the free classification task in the current study provided further evidence that the underlying structure of perceptual dialect category representations reflects important linguistic and sociolinguistic factors.
Modeling coherent errors in quantum error correction
Greenbaum, Daniel; Dutton, Zachary
2018-01-01
Analysis of quantum error correcting codes is typically done using a stochastic, Pauli channel error model for describing the noise on physical qubits. However, it was recently found that coherent errors (systematic rotations) on physical data qubits result in both physical and logical error rates that differ significantly from those predicted by a Pauli model. Here we examine the accuracy of the Pauli approximation for noise containing coherent errors (characterized by a rotation angle ɛ) under the repetition code. We derive an analytic expression for the logical error channel as a function of arbitrary code distance d and concatenation level n, in the small error limit. We find that coherent physical errors result in logical errors that are partially coherent and therefore non-Pauli. However, the coherent part of the logical error is negligible at fewer than {ε }-({dn-1)} error correction cycles when the decoder is optimized for independent Pauli errors, thus providing a regime of validity for the Pauli approximation. Above this number of correction cycles, the persistent coherent logical error will cause logical failure more quickly than the Pauli model would predict, and this may need to be combated with coherent suppression methods at the physical level or larger codes.
Semi-supervised morphosyntactic classification of Old Icelandic.
Urban, Kryztof; Tangherlini, Timothy R; Vijūnas, Aurelijus; Broadwell, Peter M
2014-01-01
We present IceMorph, a semi-supervised morphosyntactic analyzer of Old Icelandic. In addition to machine-read corpora and dictionaries, it applies a small set of declension prototypes to map corpus words to dictionary entries. A web-based GUI allows expert users to modify and augment data through an online process. A machine learning module incorporates prototype data, edit-distance metrics, and expert feedback to continuously update part-of-speech and morphosyntactic classification. An advantage of the analyzer is its ability to achieve competitive classification accuracy with minimum training data.
Robust tissue classification for reproducible wound assessment in telemedicine environments
Wannous, Hazem; Treuillet, Sylvie; Lucas, Yves
2010-04-01
In telemedicine environments, a standardized and reproducible assessment of wounds, using a simple free-handled digital camera, is an essential requirement. However, to ensure robust tissue classification, particular attention must be paid to the complete design of the color processing chain. We introduce the key steps including color correction, merging of expert labeling, and segmentation-driven classification based on support vector machines. The tool thus developed ensures stability under lighting condition, viewpoint, and camera changes, to achieve accurate and robust classification of skin tissues. Clinical tests demonstrate that such an advanced tool, which forms part of a complete 3-D and color wound assessment system, significantly improves the monitoring of the healing process. It achieves an overlap score of 79.3 against 69.1% for a single expert, after mapping on the medical reference developed from the image labeling by a college of experts.
Errors in causal inference: an organizational schema for systematic error and random error.
Suzuki, Etsuji; Tsuda, Toshihide; Mitsuhashi, Toshiharu; Mansournia, Mohammad Ali; Yamamoto, Eiji
2016-11-01
To provide an organizational schema for systematic error and random error in estimating causal measures, aimed at clarifying the concept of errors from the perspective of causal inference. We propose to divide systematic error into structural error and analytic error. With regard to random error, our schema shows its four major sources: nondeterministic counterfactuals, sampling variability, a mechanism that generates exposure events and measurement variability. Structural error is defined from the perspective of counterfactual reasoning and divided into nonexchangeability bias (which comprises confounding bias and selection bias) and measurement bias. Directed acyclic graphs are useful to illustrate this kind of error. Nonexchangeability bias implies a lack of "exchangeability" between the selected exposed and unexposed groups. A lack of exchangeability is not a primary concern of measurement bias, justifying its separation from confounding bias and selection bias. Many forms of analytic errors result from the small-sample properties of the estimator used and vanish asymptotically. Analytic error also results from wrong (misspecified) statistical models and inappropriate statistical methods. Our organizational schema is helpful for understanding the relationship between systematic error and random error from a previously less investigated aspect, enabling us to better understand the relationship between accuracy, validity, and precision. Copyright © 2016 Elsevier Inc. All rights reserved.
Training strategy for convolutional neural networks in pedestrian gender classification
Ng, Choon-Boon; Tay, Yong-Haur; Goi, Bok-Min
2017-06-01
In this work, we studied a strategy for training a convolutional neural network in pedestrian gender classification with limited amount of labeled training data. Unsupervised learning by k-means clustering on pedestrian images was used to learn the filters to initialize the first layer of the network. As a form of pre-training, supervised learning for the related task of pedestrian classification was performed. Finally, the network was fine-tuned for gender classification. We found that this strategy improved the network's generalization ability in gender classification, achieving better test results when compared to random weights initialization and slightly more beneficial than merely initializing the first layer filters by unsupervised learning. This shows that unsupervised learning followed by pre-training with pedestrian images is an effective strategy to learn useful features for pedestrian gender classification.
PHOTOMETRIC SUPERNOVA CLASSIFICATION WITH MACHINE LEARNING
Energy Technology Data Exchange (ETDEWEB)
Lochner, Michelle; Peiris, Hiranya V.; Lahav, Ofer; Winter, Max K. [Department of Physics and Astronomy, University College London, Gower Street, London WC1E 6BT (United Kingdom); McEwen, Jason D., E-mail: dr.michelle.lochner@gmail.com [Mullard Space Science Laboratory, University College London, Surrey RH5 6NT (United Kingdom)
2016-08-01
Automated photometric supernova classification has become an active area of research in recent years in light of current and upcoming imaging surveys such as the Dark Energy Survey (DES) and the Large Synoptic Survey Telescope, given that spectroscopic confirmation of type for all supernovae discovered will be impossible. Here, we develop a multi-faceted classification pipeline, combining existing and new approaches. Our pipeline consists of two stages: extracting descriptive features from the light curves and classification using a machine learning algorithm. Our feature extraction methods vary from model-dependent techniques, namely SALT2 fits, to more independent techniques that fit parametric models to curves, to a completely model-independent wavelet approach. We cover a range of representative machine learning algorithms, including naive Bayes, k -nearest neighbors, support vector machines, artificial neural networks, and boosted decision trees (BDTs). We test the pipeline on simulated multi-band DES light curves from the Supernova Photometric Classification Challenge. Using the commonly used area under the curve (AUC) of the Receiver Operating Characteristic as a metric, we find that the SALT2 fits and the wavelet approach, with the BDTs algorithm, each achieve an AUC of 0.98, where 1 represents perfect classification. We find that a representative training set is essential for good classification, whatever the feature set or algorithm, with implications for spectroscopic follow-up. Importantly, we find that by using either the SALT2 or the wavelet feature sets with a BDT algorithm, accurate classification is possible purely from light curve data, without the need for any redshift information.
PHOTOMETRIC SUPERNOVA CLASSIFICATION WITH MACHINE LEARNING
International Nuclear Information System (INIS)
Lochner, Michelle; Peiris, Hiranya V.; Lahav, Ofer; Winter, Max K.; McEwen, Jason D.
2016-01-01
Automated photometric supernova classification has become an active area of research in recent years in light of current and upcoming imaging surveys such as the Dark Energy Survey (DES) and the Large Synoptic Survey Telescope, given that spectroscopic confirmation of type for all supernovae discovered will be impossible. Here, we develop a multi-faceted classification pipeline, combining existing and new approaches. Our pipeline consists of two stages: extracting descriptive features from the light curves and classification using a machine learning algorithm. Our feature extraction methods vary from model-dependent techniques, namely SALT2 fits, to more independent techniques that fit parametric models to curves, to a completely model-independent wavelet approach. We cover a range of representative machine learning algorithms, including naive Bayes, k -nearest neighbors, support vector machines, artificial neural networks, and boosted decision trees (BDTs). We test the pipeline on simulated multi-band DES light curves from the Supernova Photometric Classification Challenge. Using the commonly used area under the curve (AUC) of the Receiver Operating Characteristic as a metric, we find that the SALT2 fits and the wavelet approach, with the BDTs algorithm, each achieve an AUC of 0.98, where 1 represents perfect classification. We find that a representative training set is essential for good classification, whatever the feature set or algorithm, with implications for spectroscopic follow-up. Importantly, we find that by using either the SALT2 or the wavelet feature sets with a BDT algorithm, accurate classification is possible purely from light curve data, without the need for any redshift information.
Abu-Alqumsan, Mohammad; Kapeller, Christoph; Hintermüller, Christoph; Guger, Christoph; Peer, Angelika
2017-12-01
Objective. This paper discusses the invariance and variability in interaction error-related potentials (ErrPs), where a special focus is laid upon the factors of (1) the human mental processing required to assess interface actions (2) time (3) subjects. Approach. Three different experiments were designed as to vary primarily with respect to the mental processes that are necessary to assess whether an interface error has occurred or not. The three experiments were carried out with 11 subjects in a repeated-measures experimental design. To study the effect of time, a subset of the recruited subjects additionally performed the same experiments on different days. Main results. The ErrP variability across the different experiments for the same subjects was found largely attributable to the different mental processing required to assess interface actions. Nonetheless, we found that interaction ErrPs are empirically invariant over time (for the same subject and same interface) and to a lesser extent across subjects (for the same interface). Significance. The obtained results may be used to explain across-study variability of ErrPs, as well as to define guidelines for approaches to the ErrP classifier transferability problem.
Digitisation of films and texture analysis for digital classification of pulmonary opacities
International Nuclear Information System (INIS)
Desaga, J.F.; Dengler, J.; Wolf, T.; Engelmann, U.; Scheppelmann, D.; Meinzer, H.P.
1988-01-01
The study aimed at evaluating the effect of different methods of digitisation of radiographic films on the digital classification of pulmonary opacities. Test sets from the standard of the International Labour Office (ILO) Classification of Radiographs of Pneumoconiosis were prepared by film digitsation using a scanning microdensitometer or a video digitiser based on a personal computer equipped with a real time digitiser board and a vidicon or a Charge Coupled Device (CCD) camera. Seven different algorithms were used for texture analysis resulting in 16 texture parameters for each region. All methods used for texture analysis were independent of the mean grey value level and the size of the image analysed. Classification was performed by discriminant analysis using the classes from the ILO classification. A hit ratio of at least 85% was achieved for a digitisation by scanner digitisation or the vidicon, while the corresponding results of the CCD camera were significantly less good. Classification by texture analysis of opacities of chest X-rays of pneumoconiosis digitised by a personal computer based video digitiser and a vidicon are of equal quality compared to digitisation by a scanning microdensitometer. Correct classification of 90% was achieved via the described statistical approach. (orig.) [de
Classification of proteins: available structural space for molecular modeling.
Andreeva, Antonina
2012-01-01
The wealth of available protein structural data provides unprecedented opportunity to study and better understand the underlying principles of protein folding and protein structure evolution. A key to achieving this lies in the ability to analyse these data and to organize them in a coherent classification scheme. Over the past years several protein classifications have been developed that aim to group proteins based on their structural relationships. Some of these classification schemes explore the concept of structural neighbourhood (structural continuum), whereas other utilize the notion of protein evolution and thus provide a discrete rather than continuum view of protein structure space. This chapter presents a strategy for classification of proteins with known three-dimensional structure. Steps in the classification process along with basic definitions are introduced. Examples illustrating some fundamental concepts of protein folding and evolution with a special focus on the exceptions to them are presented.
International Nuclear Information System (INIS)
Rand, C.P. du; Schoor, G. van
2012-01-01
Highlights: ► Different uncorrelated fault signatures are derived for HTGR component faults. ► A multiple classifier ensemble increases confidence in classification accuracy. ► Detailed simulation model of system is not required for fault diagnosis. - Abstract: The second paper in a two part series presents the area error method for generation of representative enthalpy–entropy (h–s) fault signatures to classify malfunctions in generation IV nuclear high temperature gas-cooled reactor (HTGR) components. The second classifier is devised to ultimately address the fault diagnosis (FD) problem via the proposed methods in a multiple classifier (MC) ensemble. FD is realized by way of different input feature sets to the classification algorithm based on the area and trajectory of the residual shift between the fault-free and the actual operating h–s graph models. The application of the proposed technique is specifically demonstrated for 24 single fault transients considered in the main power system (MPS) of the Pebble Bed Modular Reactor (PBMR). The results show that the area error technique produces different fault signatures with low correlation for all the examined component faults. A brief evaluation of the two fault signature generation techniques is presented and the performance of the area error method is documented using the fault classification index (FCI) presented in Part I of the series. The final part of this work reports the application of the proposed approach for classification of an emulated fault transient in data from the prototype Pebble Bed Micro Model (PBMM) plant. Reference data values are calculated for the plant via a thermo-hydraulic simulation model of the MPS. The results show that the correspondence between the fault signatures, generated via experimental plant data and simulated reference values, are generally good. The work presented in the two part series, related to the classification of component faults in the MPS of different
Classification of COROT Exoplanet Light Curves
Debosscher, J.; Aerts, C.C.; Vandenbussche, B.
2006-01-01
We present methodology to achieve the automated variability classification of stars based on photometric time series. Our work is done in the framework of the COROT space mission to be launched in 2006, but will also be applicable to data of the future Gaia satellite. We developed routines that are
A Novel Classification Method for Syndrome Differentiation of Patients with AIDS
Directory of Open Access Journals (Sweden)
Yufeng Zhao
2015-01-01
Full Text Available We consider the analysis of an AIDS dataset where each patient is characterized by a list of symptoms and is labeled with one or more TCM syndromes. The task is to build a classifier that maps symptoms to TCM syndromes. We use the minimum reference set-based multiple instance learning (MRS-MIL method. The method identifies a list of representative symptoms for each syndrome and builds a Gaussian mixture model based on them. The models for all syndromes are then used for classification via Bayes rule. By relying on a subset of key symptoms for classification, MRS-MIL can produce reliable and high quality classification rules even on datasets with small sample size. On the AIDS dataset, it achieves average precision and recall 0.7736 and 0.7111, respectively. Those are superior to results achieved by alternative methods.
Influence of Ephemeris Error on GPS Single Point Positioning Accuracy
Lihua, Ma; Wang, Meng
2013-09-01
The Global Positioning System (GPS) user makes use of the navigation message transmitted from GPS satellites to achieve its location. Because the receiver uses the satellite's location in position calculations, an ephemeris error, a difference between the expected and actual orbital position of a GPS satellite, reduces user accuracy. The influence extent is decided by the precision of broadcast ephemeris from the control station upload. Simulation analysis with the Yuma almanac show that maximum positioning error exists in the case where the ephemeris error is along the line-of-sight (LOS) direction. Meanwhile, the error is dependent on the relationship between the observer and spatial constellation at some time period.
Iris Image Classification Based on Hierarchical Visual Codebook.
Zhenan Sun; Hui Zhang; Tieniu Tan; Jianyu Wang
2014-06-01
Iris recognition as a reliable method for personal identification has been well-studied with the objective to assign the class label of each iris image to a unique subject. In contrast, iris image classification aims to classify an iris image to an application specific category, e.g., iris liveness detection (classification of genuine and fake iris images), race classification (e.g., classification of iris images of Asian and non-Asian subjects), coarse-to-fine iris identification (classification of all iris images in the central database into multiple categories). This paper proposes a general framework for iris image classification based on texture analysis. A novel texture pattern representation method called Hierarchical Visual Codebook (HVC) is proposed to encode the texture primitives of iris images. The proposed HVC method is an integration of two existing Bag-of-Words models, namely Vocabulary Tree (VT), and Locality-constrained Linear Coding (LLC). The HVC adopts a coarse-to-fine visual coding strategy and takes advantages of both VT and LLC for accurate and sparse representation of iris texture. Extensive experimental results demonstrate that the proposed iris image classification method achieves state-of-the-art performance for iris liveness detection, race classification, and coarse-to-fine iris identification. A comprehensive fake iris image database simulating four types of iris spoof attacks is developed as the benchmark for research of iris liveness detection.
NEW CLASSIFICATION OF ECOPOLICES
Directory of Open Access Journals (Sweden)
VOROBYOV V. V.
2016-09-01
Full Text Available Problem statement. Ecopolices are the newest stage of the urban planning. They have to be consideredsuchas material and energy informational structures, included to the dynamic-evolutionary matrix netsofex change processes in the ecosystems. However, there are not made the ecopolice classifications, developing on suchapproaches basis. And this determined the topicality of the article. Analysis of publications on theoretical and applied aspects of the ecopolices formation showed, that the work on them is managed mainly in the context of the latest scientific and technological achievements in the various knowledge fields. These settlements are technocratic. They are connected with the morphology of space, network structures of regional and local natural ecosystems, without independent stability, can not exist without continuous man support. Another words, they do not work in with an ecopolices idea. It is come to a head for objective, symbiotic searching of ecopolices concept with the development of their classifications. Purpose statement is to develop the objective evidence for ecopolices and to propose their new classification. Conclusion. On the base of the ecopolices classification have to lie an elements correlation idea of their general plans and men activity type according with natural mechanism of accepting, reworking and transmission of material, energy and information between geo-ecosystems, planet, man, ecopolices material part and Cosmos. New ecopolices classification should be based on the principles of multi-dimensional, time-spaced symbiotic clarity with exchange ecosystem networks. The ecopolice function with this approach comes not from the subjective anthropocentric economy but from the holistic objective of Genesis paradigm. Or, otherwise - not from the Consequence, but from the Cause.
Computational error and complexity in science and engineering computational error and complexity
Lakshmikantham, Vangipuram; Chui, Charles K; Chui, Charles K
2005-01-01
The book "Computational Error and Complexity in Science and Engineering pervades all the science and engineering disciplines where computation occurs. Scientific and engineering computation happens to be the interface between the mathematical model/problem and the real world application. One needs to obtain good quality numerical values for any real-world implementation. Just mathematical quantities symbols are of no use to engineers/technologists. Computational complexity of the numerical method to solve the mathematical model, also computed along with the solution, on the other hand, will tell us how much computation/computational effort has been spent to achieve that quality of result. Anyone who wants the specified physical problem to be solved has every right to know the quality of the solution as well as the resources spent for the solution. The computed error as well as the complexity provide the scientific convincing answer to these questions. Specifically some of the disciplines in which the book w...
Overfitting Reduction of Text Classification Based on AdaBELM
Directory of Open Access Journals (Sweden)
Xiaoyue Feng
2017-07-01
Full Text Available Overfitting is an important problem in machine learning. Several algorithms, such as the extreme learning machine (ELM, suffer from this issue when facing high-dimensional sparse data, e.g., in text classification. One common issue is that the extent of overfitting is not well quantified. In this paper, we propose a quantitative measure of overfitting referred to as the rate of overfitting (RO and a novel model, named AdaBELM, to reduce the overfitting. With RO, the overfitting problem can be quantitatively measured and identified. The newly proposed model can achieve high performance on multi-class text classification. To evaluate the generalizability of the new model, we designed experiments based on three datasets, i.e., the 20 Newsgroups, Reuters-21578, and BioMed corpora, which represent balanced, unbalanced, and real application data, respectively. Experiment results demonstrate that AdaBELM can reduce overfitting and outperform classical ELM, decision tree, random forests, and AdaBoost on all three text-classification datasets; for example, it can achieve 62.2% higher accuracy than ELM. Therefore, the proposed model has a good generalizability.
Chinese Sentence Classification Based on Convolutional Neural Network
Gu, Chengwei; Wu, Ming; Zhang, Chuang
2017-10-01
Sentence classification is one of the significant issues in Natural Language Processing (NLP). Feature extraction is often regarded as the key point for natural language processing. Traditional ways based on machine learning can not take high level features into consideration, such as Naive Bayesian Model. The neural network for sentence classification can make use of contextual information to achieve greater results in sentence classification tasks. In this paper, we focus on classifying Chinese sentences. And the most important is that we post a novel architecture of Convolutional Neural Network (CNN) to apply on Chinese sentence classification. In particular, most of the previous methods often use softmax classifier for prediction, we embed a linear support vector machine to substitute softmax in the deep neural network model, minimizing a margin-based loss to get a better result. And we use tanh as an activation function, instead of ReLU. The CNN model improve the result of Chinese sentence classification tasks. Experimental results on the Chinese news title database validate the effectiveness of our model.
Directory of Open Access Journals (Sweden)
Muhammad Ahmad
Full Text Available Hyperspectral image classification with a limited number of training samples without loss of accuracy is desirable, as collecting such data is often expensive and time-consuming. However, classifiers trained with limited samples usually end up with a large generalization error. To overcome the said problem, we propose a fuzziness-based active learning framework (FALF, in which we implement the idea of selecting optimal training samples to enhance generalization performance for two different kinds of classifiers, discriminative and generative (e.g. SVM and KNN. The optimal samples are selected by first estimating the boundary of each class and then calculating the fuzziness-based distance between each sample and the estimated class boundaries. Those samples that are at smaller distances from the boundaries and have higher fuzziness are chosen as target candidates for the training set. Through detailed experimentation on three publically available datasets, we showed that when trained with the proposed sample selection framework, both classifiers achieved higher classification accuracy and lower processing time with the small amount of training data as opposed to the case where the training samples were selected randomly. Our experiments demonstrate the effectiveness of our proposed method, which equates favorably with the state-of-the-art methods.
Error floor behavior study of LDPC codes for concatenated codes design
Chen, Weigang; Yin, Liuguo; Lu, Jianhua
2007-11-01
Error floor behavior of low-density parity-check (LDPC) codes using quantized decoding algorithms is statistically studied with experimental results on a hardware evaluation platform. The results present the distribution of the residual errors after decoding failure and reveal that the number of residual error bits in a codeword is usually very small using quantized sum-product (SP) algorithm. Therefore, LDPC code may serve as the inner code in a concatenated coding system with a high code rate outer code and thus an ultra low error floor can be achieved. This conclusion is also verified by the experimental results.
Enhanced Named Entity Extraction via Error-Driven Aggregation
Energy Technology Data Exchange (ETDEWEB)
Lemmond, T D; Perry, N C; Guensche, J W; Nitao, J J; Glaser, R E; Kidwell, P; Hanley, W G
2010-02-22
Despite recent advances in named entity extraction technologies, state-of-the-art extraction tools achieve insufficient accuracy rates for practical use in many operational settings. However, they are not generally prone to the same types of error, suggesting that substantial improvements may be achieved via appropriate combinations of existing tools, provided their behavior can be accurately characterized and quantified. In this paper, we present an inference methodology for the aggregation of named entity extraction technologies that is founded upon a black-box analysis of their respective error processes. This method has been shown to produce statistically significant improvements in extraction relative to standard performance metrics and to mitigate the weak performance of entity extractors operating under suboptimal conditions. Moreover, this approach provides a framework for quantifying uncertainty and has demonstrated the ability to reconstruct the truth when majority voting fails.
Classification of DNA nucleotides with transverse tunneling currents
DEFF Research Database (Denmark)
Pedersen, Jonas Nyvold; Boynton, Paul; Ventra, Massimiliano Di
2017-01-01
, however. In realistic liquid environments, typical currents in tunneling devices are of the order of picoamps. This corresponds to only six electrons per microsecond, and this number affects the integration time required to do current measurements in real experiments. This limits the speed of sequencing......, though current fluctuations due to Brownian motion of the molecule average out during the required integration time. Moreover, data acquisition equipment introduces noise, and electronic filters create correlations in time-series data. We discuss how these effects must be included in the analysis of, e.......g., the assignment of specific nucleobases to current signals. As the signals from different molecules overlap, unambiguous classification is impossible with a single measurement. We argue that the assignment of molecules to a signal is a standard pattern classification problem and calculation of the error rates...
A posteriori error estimator and AMR for discrete ordinates nodal transport methods
International Nuclear Information System (INIS)
Duo, Jose I.; Azmy, Yousry Y.; Zikatanov, Ludmil T.
2009-01-01
In the development of high fidelity transport solvers, optimization of the use of available computational resources and access to a tool for assessing quality of the solution are key to the success of large-scale nuclear systems' simulation. In this regard, error control provides the analyst with a confidence level in the numerical solution and enables for optimization of resources through Adaptive Mesh Refinement (AMR). In this paper, we derive an a posteriori error estimator based on the nodal solution of the Arbitrarily High Order Transport Method of the Nodal type (AHOT-N). Furthermore, by making assumptions on the regularity of the solution, we represent the error estimator as a function of computable volume and element-edges residuals. The global L 2 error norm is proved to be bound by the estimator. To lighten the computational load, we present a numerical approximation to the aforementioned residuals and split the global norm error estimator into local error indicators. These indicators are used to drive an AMR strategy for the spatial discretization. However, the indicators based on forward solution residuals alone do not bound the cell-wise error. The estimator and AMR strategy are tested in two problems featuring strong heterogeneity and highly transport streaming regime with strong flux gradients. The results show that the error estimator indeed bounds the global error norms and that the error indicator follows the cell-error's spatial distribution pattern closely. The AMR strategy proves beneficial to optimize resources, primarily by reducing the number of unknowns solved for to achieve prescribed solution accuracy in global L 2 error norm. Likewise, AMR achieves higher accuracy compared to uniform refinement when resolving sharp flux gradients, for the same number of unknowns
Error begat error: design error analysis and prevention in social infrastructure projects.
Love, Peter E D; Lopez, Robert; Edwards, David J; Goh, Yang M
2012-09-01
Design errors contribute significantly to cost and schedule growth in social infrastructure projects and to engineering failures, which can result in accidents and loss of life. Despite considerable research that has addressed their error causation in construction projects they still remain prevalent. This paper identifies the underlying conditions that contribute to design errors in social infrastructure projects (e.g. hospitals, education, law and order type buildings). A systemic model of error causation is propagated and subsequently used to develop a learning framework for design error prevention. The research suggests that a multitude of strategies should be adopted in congruence to prevent design errors from occurring and so ensure that safety and project performance are ameliorated. Copyright © 2011. Published by Elsevier Ltd.
Pornography classification: The hidden clues in video space-time.
Moreira, Daniel; Avila, Sandra; Perez, Mauricio; Moraes, Daniel; Testoni, Vanessa; Valle, Eduardo; Goldenstein, Siome; Rocha, Anderson
2016-11-01
As web technologies and social networks become part of the general public's life, the problem of automatically detecting pornography is into every parent's mind - nobody feels completely safe when their children go online. In this paper, we focus on video-pornography classification, a hard problem in which traditional methods often employ still-image techniques - labeling frames individually prior to a global decision. Frame-based approaches, however, ignore significant cogent information brought by motion. Here, we introduce a space-temporal interest point detector and descriptor called Temporal Robust Features (TRoF). TRoF was custom-tailored for efficient (low processing time and memory footprint) and effective (high classification accuracy and low false negative rate) motion description, particularly suited to the task at hand. We aggregate local information extracted by TRoF into a mid-level representation using Fisher Vectors, the state-of-the-art model of Bags of Visual Words (BoVW). We evaluate our original strategy, contrasting it both to commercial pornography detection solutions, and to BoVW solutions based upon other space-temporal features from the scientific literature. The performance is assessed using the Pornography-2k dataset, a new challenging pornographic benchmark, comprising 2000 web videos and 140h of video footage. The dataset is also a contribution of this work and is very assorted, including both professional and amateur content, and it depicts several genres of pornography, from cartoon to live action, with diverse behavior and ethnicity. The best approach, based on a dense application of TRoF, yields a classification error reduction of almost 79% when compared to the best commercial classifier. A sparse description relying on TRoF detector is also noteworthy, for yielding a classification error reduction of over 69%, with 19× less memory footprint than the dense solution, and yet can also be implemented to meet real-time requirements
Out-of-Sample Generalizations for Supervised Manifold Learning for Classification.
Vural, Elif; Guillemot, Christine
2016-03-01
Supervised manifold learning methods for data classification map high-dimensional data samples to a lower dimensional domain in a structure-preserving way while increasing the separation between different classes. Most manifold learning methods compute the embedding only of the initially available data; however, the generalization of the embedding to novel points, i.e., the out-of-sample extension problem, becomes especially important in classification applications. In this paper, we propose a semi-supervised method for building an interpolation function that provides an out-of-sample extension for general supervised manifold learning algorithms studied in the context of classification. The proposed algorithm computes a radial basis function interpolator that minimizes an objective function consisting of the total embedding error of unlabeled test samples, defined as their distance to the embeddings of the manifolds of their own class, as well as a regularization term that controls the smoothness of the interpolation function in a direction-dependent way. The class labels of test data and the interpolation function parameters are estimated jointly with an iterative process. Experimental results on face and object images demonstrate the potential of the proposed out-of-sample extension algorithm for the classification of manifold-modeled data sets.
Vehicle Classification and Speed Estimation Using Combined Passive Infrared/Ultrasonic Sensors
Odat, Enas M.
2017-09-18
In this paper, a new sensing device that can simultaneously monitor traffic congestion and urban flash floods is presented. This sensing device is based on the combination of passive infrared sensors (PIRs) and ultrasonic rangefinder, and is used for real-time vehicle detection, classification, and speed estimation in the context of wireless sensor networks. This framework relies on dynamic Bayesian Networks to fuse heterogeneous data both spatially and temporally for vehicle detection. To estimate the speed of the incoming vehicles, we first use cross correlation and wavelet transform-based methods to estimate the time delay between the signals of different sensors. We then propose a calibration and self-correction model based on Bayesian Networks to make a joint inference by all sensors about the speed and the length of the detected vehicle. Furthermore, we use the measurements of the ultrasonic and the PIR sensors to perform vehicle classification. Validation data (using an experimental dual infrared and ultrasonic traffic sensor) show a 99% accuracy in vehicle detection, a mean error of 5 kph in vehicle speed estimation, a mean error of 0.7m in vehicle length estimation, and a high accuracy in vehicle classification. Finally, we discuss the computational performance of the algorithm, and show that this framework can be implemented on low-power computational devices within a wireless sensor network setting. Such decentralized processing greatly improves the energy consumption of the system and minimizes bandwidth usage.
Susanti, Yuliana; Zukhronah, Etik; Pratiwi, Hasih; Respatiwulan; Sri Sulistijowati, H.
2017-11-01
To achieve food resilience in Indonesia, food diversification by exploring potentials of local food is required. Corn is one of alternating staple food of Javanese society. For that reason, corn production needs to be improved by considering the influencing factors. CHAID and CRT are methods of data mining which can be used to classify the influencing variables. The present study seeks to dig up information on the potentials of local food availability of corn in regencies and cities in Java Island. CHAID analysis yields four classifications with accuracy of 78.8%, while CRT analysis yields seven classifications with accuracy of 79.6%.
Low-cost real-time automatic wheel classification system
Shabestari, Behrouz N.; Miller, John W. V.; Wedding, Victoria
1992-11-01
This paper describes the design and implementation of a low-cost machine vision system for identifying various types of automotive wheels which are manufactured in several styles and sizes. In this application, a variety of wheels travel on a conveyor in random order through a number of processing steps. One of these processes requires the identification of the wheel type which was performed manually by an operator. A vision system was designed to provide the required identification. The system consisted of an annular illumination source, a CCD TV camera, frame grabber, and 386-compatible computer. Statistical pattern recognition techniques were used to provide robust classification as well as a simple means for adding new wheel designs to the system. Maintenance of the system can be performed by plant personnel with minimal training. The basic steps for identification include image acquisition, segmentation of the regions of interest, extraction of selected features, and classification. The vision system has been installed in a plant and has proven to be extremely effective. The system properly identifies the wheels correctly up to 30 wheels per minute regardless of rotational orientation in the camera's field of view. Correct classification can even be achieved if a portion of the wheel is blocked off from the camera. Significant cost savings have been achieved by a reduction in scrap associated with incorrect manual classification as well as a reduction of labor in a tedious task.
Inventory classification based on decoupling points
Directory of Open Access Journals (Sweden)
Joakim Wikner
2015-01-01
Full Text Available The ideal state of continuous one-piece flow may never be achieved. Still the logistics manager can improve the flow by carefully positioning inventory to buffer against variations. Strategies such as lean, postponement, mass customization, and outsourcing all rely on strategic positioning of decoupling points to separate forecast-driven from customer-order-driven flows. Planning and scheduling of the flow are also based on classification of decoupling points as master scheduled or not. A comprehensive classification scheme for these types of decoupling points is introduced. The approach rests on identification of flows as being either demand based or supply based. The demand or supply is then combined with exogenous factors, classified as independent, or endogenous factors, classified as dependent. As a result, eight types of strategic as well as tactical decoupling points are identified resulting in a process-based framework for inventory classification that can be used for flow design.
DEFF Research Database (Denmark)
Debus, Michael S.
2017-01-01
This paper critically analyzes seventeen game classifications. The classifications were chosen on the basis of diversity, ranging from pre-digital classification (e.g. Murray 1952), over game studies classifications (e.g. Elverdam & Aarseth 2007) to classifications of drinking games (e.g. LaBrie et...... al. 2013). The analysis aims at three goals: The classifications’ internal consistency, the abstraction of classification criteria and the identification of differences in classification across fields and/or time. Especially the abstraction of classification criteria can be used in future endeavors...... into the topic of game classifications....
Supervised learning for the automated transcription of spacer classification from spoligotype films
Directory of Open Access Journals (Sweden)
Abernethy Neil
2009-08-01
Full Text Available Abstract Background Molecular genotyping of bacteria has revolutionized the study of tuberculosis epidemiology, yet these established laboratory techniques typically require subjective and laborious interpretation by trained professionals. In the context of a Tuberculosis Case Contact study in The Gambia we used a reverse hybridization laboratory assay called spoligotype analysis. To facilitate processing of spoligotype images we have developed tools and algorithms to automate the classification and transcription of these data directly to a database while allowing for manual editing. Results Features extracted from each of the 1849 spots on a spoligo film were classified using two supervised learning algorithms. A graphical user interface allows manual editing of the classification, before export to a database. The application was tested on ten films of differing quality and the results of the best classifier were compared to expert manual classification, giving a median correct classification rate of 98.1% (inter quartile range: 97.1% to 99.2%, with an automated processing time of less than 1 minute per film. Conclusion The software implementation offers considerable time savings over manual processing whilst allowing expert editing of the automated classification. The automatic upload of the classification to a database reduces the chances of transcription errors.
Discriminative clustering on manifold for adaptive transductive classification.
Zhang, Zhao; Jia, Lei; Zhang, Min; Li, Bing; Zhang, Li; Li, Fanzhang
2017-10-01
In this paper, we mainly propose a novel adaptive transductive label propagation approach by joint discriminative clustering on manifolds for representing and classifying high-dimensional data. Our framework seamlessly combines the unsupervised manifold learning, discriminative clustering and adaptive classification into a unified model. Also, our method incorporates the adaptive graph weight construction with label propagation. Specifically, our method is capable of propagating label information using adaptive weights over low-dimensional manifold features, which is different from most existing studies that usually predict the labels and construct the weights in the original Euclidean space. For transductive classification by our formulation, we first perform the joint discriminative K-means clustering and manifold learning to capture the low-dimensional nonlinear manifolds. Then, we construct the adaptive weights over the learnt manifold features, where the adaptive weights are calculated through performing the joint minimization of the reconstruction errors over features and soft labels so that the graph weights can be joint-optimal for data representation and classification. Using the adaptive weights, we can easily estimate the unknown labels of samples. After that, our method returns the updated weights for further updating the manifold features. Extensive simulations on image classification and segmentation show that our proposed algorithm can deliver the state-of-the-art performance on several public datasets. Copyright © 2017 Elsevier Ltd. All rights reserved.
An Automated and Intelligent Medical Decision Support System for Brain MRI Scans Classification.
Directory of Open Access Journals (Sweden)
Muhammad Faisal Siddiqui
Full Text Available A wide interest has been observed in the medical health care applications that interpret neuroimaging scans by machine learning systems. This research proposes an intelligent, automatic, accurate, and robust classification technique to classify the human brain magnetic resonance image (MRI as normal or abnormal, to cater down the human error during identifying the diseases in brain MRIs. In this study, fast discrete wavelet transform (DWT, principal component analysis (PCA, and least squares support vector machine (LS-SVM are used as basic components. Firstly, fast DWT is employed to extract the salient features of brain MRI, followed by PCA, which reduces the dimensions of the features. These reduced feature vectors also shrink the memory storage consumption by 99.5%. At last, an advanced classification technique based on LS-SVM is applied to brain MR image classification using reduced features. For improving the efficiency, LS-SVM is used with non-linear radial basis function (RBF kernel. The proposed algorithm intelligently determines the optimized values of the hyper-parameters of the RBF kernel and also applied k-fold stratified cross validation to enhance the generalization of the system. The method was tested by 340 patients' benchmark datasets of T1-weighted and T2-weighted scans. From the analysis of experimental results and performance comparisons, it is observed that the proposed medical decision support system outperformed all other modern classifiers and achieves 100% accuracy rate (specificity/sensitivity 100%/100%. Furthermore, in terms of computation time, the proposed technique is significantly faster than the recent well-known methods, and it improves the efficiency by 71%, 3%, and 4% on feature extraction stage, feature reduction stage, and classification stage, respectively. These results indicate that the proposed well-trained machine learning system has the potential to make accurate predictions about brain abnormalities
Automatic liver volume segmentation and fibrosis classification
Bal, Evgeny; Klang, Eyal; Amitai, Michal; Greenspan, Hayit
2018-02-01
In this work, we present an automatic method for liver segmentation and fibrosis classification in liver computed-tomography (CT) portal phase scans. The input is a full abdomen CT scan with an unknown number of slices, and the output is a liver volume segmentation mask and a fibrosis grade. A multi-stage analysis scheme is applied to each scan, including: volume segmentation, texture features extraction and SVM based classification. Data contains portal phase CT examinations from 80 patients, taken with different scanners. Each examination has a matching Fibroscan grade. The dataset was subdivided into two groups: first group contains healthy cases and mild fibrosis, second group contains moderate fibrosis, severe fibrosis and cirrhosis. Using our automated algorithm, we achieved an average dice index of 0.93 ± 0.05 for segmentation and a sensitivity of 0.92 and specificity of 0.81for classification. To the best of our knowledge, this is a first end to end automatic framework for liver fibrosis classification; an approach that, once validated, can have a great potential value in the clinic.
A Pruning Neural Network Model in Credit Classification Analysis
Directory of Open Access Journals (Sweden)
Yajiao Tang
2018-01-01
Full Text Available Nowadays, credit classification models are widely applied because they can help financial decision-makers to handle credit classification issues. Among them, artificial neural networks (ANNs have been widely accepted as the convincing methods in the credit industry. In this paper, we propose a pruning neural network (PNN and apply it to solve credit classification problem by adopting the well-known Australian and Japanese credit datasets. The model is inspired by synaptic nonlinearity of a dendritic tree in a biological neural model. And it is trained by an error back-propagation algorithm. The model is capable of realizing a neuronal pruning function by removing the superfluous synapses and useless dendrites and forms a tidy dendritic morphology at the end of learning. Furthermore, we utilize logic circuits (LCs to simulate the dendritic structures successfully which makes PNN be implemented on the hardware effectively. The statistical results of our experiments have verified that PNN obtains superior performance in comparison with other classical algorithms in terms of accuracy and computational efficiency.
Classification of Breast Cancer Subtypes by combining Gene Expression and DNA Methylation Data
Directory of Open Access Journals (Sweden)
List Markus
2014-06-01
Full Text Available Selecting the most promising treatment strategy for breast cancer crucially depends on determining the correct subtype. In recent years, gene expression profiling has been investigated as an alternative to histochemical methods. Since databases like TCGA provide easy and unrestricted access to gene expression data for hundreds of patients, the challenge is to extract a minimal optimal set of genes with good prognostic properties from a large bulk of genes making a moderate contribution to classification. Several studies have successfully applied machine learning algorithms to solve this so-called gene selection problem. However, more diverse data from other OMICS technologies are available, including methylation. We hypothesize that combining methylation and gene expression data could already lead to a largely improved classification model, since the resulting model will reflect differences not only on the transcriptomic, but also on an epigenetic level. We compared so-called random forest derived classification models based on gene expression and methylation data alone, to a model based on the combined features and to a model based on the gold standard PAM50. We obtained bootstrap errors of 10-20% and classification error of 1-50%, depending on breast cancer subtype and model. The gene expression model was clearly superior to the methylation model, which was also reflected in the combined model, which mainly selected features from gene expression data. However, the methylation model was able to identify unique features not considered as relevant by the gene expression model, which might provide deeper insights into breast cancer subtype differentiation on an epigenetic level.
DEFF Research Database (Denmark)
Shrestha, Santosh; Deleuran, Lise Christina; Gislum, René
2016-01-01
nm were extracted from multispectral images of tomato seeds. Principal component analysis (PCA) was used for data exploration, while partial least squares discriminant analysis (PLS-DA) and support vector machine discriminant analysis (SVM-DA) were used to classify the five different tomato cultivars....... The results showed very good classification accuracy for two independent test sets ranging from 94% to 100% for all tomato cultivars irrespective of chemometric methods. The overall classification error rates were 3.2% and 0.4% for the PLS-DA and SVM-DA calibration models, respectively. The results indicate...
SQL based cardiovascular ultrasound image classification.
Nandagopalan, S; Suryanarayana, Adiga B; Sudarshan, T S B; Chandrashekar, Dhanalakshmi; Manjunath, C N
2013-01-01
This paper proposes a novel method to analyze and classify the cardiovascular ultrasound echocardiographic images using Naïve-Bayesian model via database OLAP-SQL. Efficient data mining algorithms based on tightly-coupled model is used to extract features. Three algorithms are proposed for classification namely Naïve-Bayesian Classifier for Discrete variables (NBCD) with SQL, NBCD with OLAP-SQL, and Naïve-Bayesian Classifier for Continuous variables (NBCC) using OLAP-SQL. The proposed model is trained with 207 patient images containing normal and abnormal categories. Out of the three proposed algorithms, a high classification accuracy of 96.59% was achieved from NBCC which is better than the earlier methods.
Directory of Open Access Journals (Sweden)
Vasilisa Muntean
2017-10-01
Full Text Available In the research, a doctrinal and legal analysis of the concept of legal error is carried out. The author provides a self-defined definition of the concept addressed and highlights the main causes and conditions for the occurrence of judicial errors. At present, in the specialized legal doctrine of the Republic of Moldova, the problem of defining the judicial error has been little approached. In this respect, this scientific article is a scientific approach aimed at elucidating the theoretical and normative deficiencies and errors that occur in the area of reparation of the prejudice caused by judicial errors. In order to achieve our goal, we aim to create a core of ideas and referral mechanisms that ensure a certain interpretative and decisional homogeneity in the doctrinal and legal characterization of the phrase "judicial error".
Classification between normal and tumor tissues based on the pair-wise gene expression ratio
International Nuclear Information System (INIS)
Yap, YeeLeng; Zhang, XueWu; Ling, MT; Wang, XiangHong; Wong, YC; Danchin, Antoine
2004-01-01
Precise classification of cancer types is critically important for early cancer diagnosis and treatment. Numerous efforts have been made to use gene expression profiles to improve precision of tumor classification. However, reliable cancer-related signals are generally lacking. Using recent datasets on colon and prostate cancer, a data transformation procedure from single gene expression to pair-wise gene expression ratio is proposed. Making use of the internal consistency of each expression profiling dataset this transformation improves the signal to noise ratio of the dataset and uncovers new relevant cancer-related signals (features). The efficiency in using the transformed dataset to perform normal/tumor classification was investigated using feature partitioning with informative features (gene annotation) as discriminating axes (single gene expression or pair-wise gene expression ratio). Classification results were compared to the original datasets for up to 10-feature model classifiers. 82 and 262 genes that have high correlation to tissue phenotype were selected from the colon and prostate datasets respectively. Remarkably, data transformation of the highly noisy expression data successfully led to lower the coefficient of variation (CV) for the within-class samples as well as improved the correlation with tissue phenotypes. The transformed dataset exhibited lower CV when compared to that of single gene expression. In the colon cancer set, the minimum CV decreased from 45.3% to 16.5%. In prostate cancer, comparable CV was achieved with and without transformation. This improvement in CV, coupled with the improved correlation between the pair-wise gene expression ratio and tissue phenotypes, yielded higher classification efficiency, especially with the colon dataset – from 87.1% to 93.5%. Over 90% of the top ten discriminating axes in both datasets showed significant improvement after data transformation. The high classification efficiency achieved suggested
A Novel Vehicle Classification Using Embedded Strain Gauge Sensors
Directory of Open Access Journals (Sweden)
Qi Wang
2008-11-01
Full Text Available Abstract: This paper presents a new vehicle classification and develops a traffic monitoring detector to provide reliable vehicle classification to aid traffic management systems. The basic principle of this approach is based on measuring the dynamic strain caused by vehicles across pavement to obtain the corresponding vehicle parameters – wheelbase and number of axles – to then accurately classify the vehicle. A system prototype with five embedded strain sensors was developed to validate the accuracy and effectiveness of the classification method. According to the special arrangement of the sensors and the different time a vehicle arrived at the sensors one can estimate the vehicle’s speed accurately, corresponding to the estimated vehicle wheelbase and number of axles. Because of measurement errors and vehicle characteristics, there is a lot of overlap between vehicle wheelbase patterns. Therefore, directly setting up a fixed threshold for vehicle classification often leads to low-accuracy results. Using the machine learning pattern recognition method to deal with this problem is believed as one of the most effective tools. In this study, support vector machines (SVMs were used to integrate the classification features extracted from the strain sensors to automatically classify vehicles into five types, ranging from small vehicles to combination trucks, along the lines of the Federal Highway Administration vehicle classification guide. Test bench and field experiments will be introduced in this paper. Two support vector machines classification algorithms (one-against-all, one-against-one are used to classify single sensor data and multiple sensor combination data. Comparison of the two classification method results shows that the classification accuracy is very close using single data or multiple data. Our results indicate that using multiclass SVM-based fusion multiple sensor data significantly improves
A New Insight into Land Use Classification Based on Aggregated Mobile Phone Data
Pei, Tao; Sobolevsky, Stanislav; Ratti, Carlo; Shaw, Shih-Lung; Zhou, Chenghu
2013-01-01
Land use classification is essential for urban planning. Urban land use types can be differentiated either by their physical characteristics (such as reflectivity and texture) or social functions. Remote sensing techniques have been recognized as a vital method for urban land use classification because of their ability to capture the physical characteristics of land use. Although significant progress has been achieved in remote sensing methods designed for urban land use classification, most ...
Tanks for liquids: calibration and errors assessment
International Nuclear Information System (INIS)
Espejo, J.M.; Gutierrez Fernandez, J.; Ortiz, J.
1980-01-01
After a brief reference to some of the problems raised by tanks calibration, two methods, theoretical and experimental are presented, so as to achieve it taking into account measurement errors. The method is applied to the transfer of liquid from one tank to another. Further, a practical example is developed. (author)
Classification of operational characteristics of commercial cup-anemometers
Energy Technology Data Exchange (ETDEWEB)
Friis Pedersen, T; Schmidt Paulsen, U [Risoe National Lab., Wind Energy and Atmospheric Physics Dept., Roskilde (Denmark)
1999-03-01
The present classification of cup-anemometers is based on a procedure for classification of operational characteristics of cup-anemometers that was proposed at the EWEC `97 conference in Dublin 1997. Three definitions of wind speed are considered. The average longitudinal wind speed (ID), the average horizontal wind speed (2D) and the average vector wind speed (3D). The classification is provided in these terms, and additionally, the turbulence intensities, which are defined from the same wind speed definitions. The commercial cup-anemometers have all been calibrated in wind tunnel for the normal calibrations and angular characteristics. Friction was measured by blywheel testing, where the surrounding temperatures were varied over a wide range. The characteristics of the cup-anemometers have been fitted to the heuristic dynamic model, and the response has been calculated in time domain for prescribed ranges of external operational conditions. The results are presented in ranges of maximum deviations of `measured` average wind speed. For each definition of wind speed and turbulence intensity, the cup-anemometers are ranked according to the most precise instrument. Finally, the most important systematic error sources are commented. (au)
Classification of Breast Cancer Subtypes by combining Gene Expression and DNA Methylation Data
DEFF Research Database (Denmark)
List, Markus; Hauschild, Anne-Christin; Tan, Qihua
2014-01-01
expression data for hundreds of patients, the challenge is to extract a minimal optimal set of genes with good prognostic properties from a large bulk of genes making a moderate contribution to classification. Several studies have successfully applied machine learning algorithms to solve this so-called gene...... on the transcriptomic, but also on an epigenetic level. We compared so-called random forest derived classification models based on gene expression and methylation data alone, to a model based on the combined features and to a model based on the gold standard PAM50. We obtained bootstrap errors of 10...
Knoefel, Patrick; Loew, Fabian; Conrad, Christopher
2015-04-01
Crop maps based on classification of remotely sensed data are of increased attendance in agricultural management. This induces a more detailed knowledge about the reliability of such spatial information. However, classification of agricultural land use is often limited by high spectral similarities of the studied crop types. More, spatially and temporally varying agro-ecological conditions can introduce confusion in crop mapping. Classification errors in crop maps in turn may have influence on model outputs, like agricultural production monitoring. One major goal of the PhenoS project ("Phenological structuring to determine optimal acquisition dates for Sentinel-2 data for field crop classification"), is the detection of optimal phenological time windows for land cover classification purposes. Since many crop species are spectrally highly similar, accurate classification requires the right selection of satellite images for a certain classification task. In the course of one growing season, phenological phases exist where crops are separable with higher accuracies. For this purpose, coupling of multi-temporal spectral characteristics and phenological events is promising. The focus of this study is set on the separation of spectrally similar cereal crops like winter wheat, barley, and rye of two test sites in Germany called "Harz/Central German Lowland" and "Demmin". However, this study uses object based random forest (RF) classification to investigate the impact of image acquisition frequency and timing on crop classification uncertainty by permuting all possible combinations of available RapidEye time series recorded on the test sites between 2010 and 2014. The permutations were applied to different segmentation parameters. Then, classification uncertainty was assessed and analysed, based on the probabilistic soft-output from the RF algorithm at the per-field basis. From this soft output, entropy was calculated as a spatial measure of classification uncertainty
Optimizing Multiple Kernel Learning for the Classification of UAV Data
Directory of Open Access Journals (Sweden)
Caroline M. Gevaert
2016-12-01
Full Text Available Unmanned Aerial Vehicles (UAVs are capable of providing high-quality orthoimagery and 3D information in the form of point clouds at a relatively low cost. Their increasing popularity stresses the necessity of understanding which algorithms are especially suited for processing the data obtained from UAVs. The features that are extracted from the point cloud and imagery have different statistical characteristics and can be considered as heterogeneous, which motivates the use of Multiple Kernel Learning (MKL for classification problems. In this paper, we illustrate the utility of applying MKL for the classification of heterogeneous features obtained from UAV data through a case study of an informal settlement in Kigali, Rwanda. Results indicate that MKL can achieve a classification accuracy of 90.6%, a 5.2% increase over a standard single-kernel Support Vector Machine (SVM. A comparison of seven MKL methods indicates that linearly-weighted kernel combinations based on simple heuristics are competitive with respect to computationally-complex, non-linear kernel combination methods. We further underline the importance of utilizing appropriate feature grouping strategies for MKL, which has not been directly addressed in the literature, and we propose a novel, automated feature grouping method that achieves a high classification accuracy for various MKL methods.
Energy Efficient Error-Correcting Coding for Wireless Systems
Shao, X.
2010-01-01
The wireless channel is a hostile environment. The transmitted signal does not only suffers multi-path fading but also noise and interference from other users of the wireless channel. That causes unreliable communications. To achieve high-quality communications, error correcting coding is required
Zhang, Yang; Liu, Yun; Chao, Han-Chieh; Zhang, Zhenjiang; Zhang, Zhiyuan
2018-03-30
In wireless sensor networks, the classification of incomplete data reported by sensor nodes is an open issue because it is difficult to accurately estimate the missing values. In many cases, the misclassification is unacceptable considering that it probably brings catastrophic damages to the data users. In this paper, a novel classification approach of incomplete data is proposed to reduce the misclassification errors. This method uses the regularized extreme learning machine to estimate the potential values of missing data at first, and then it converts the estimations into multiple classification results on the basis of the distance between interval numbers. Finally, an evidential reasoning rule is adopted to fuse these classification results. The final decision is made according to the combined basic belief assignment. The experimental results show that this method has better performance than other traditional classification methods of incomplete data.
Raster Vs. Point Cloud LiDAR Data Classification
El-Ashmawy, N.; Shaker, A.
2014-09-01
Airborne Laser Scanning systems with light detection and ranging (LiDAR) technology is one of the fast and accurate 3D point data acquisition techniques. Generating accurate digital terrain and/or surface models (DTM/DSM) is the main application of collecting LiDAR range data. Recently, LiDAR range and intensity data have been used for land cover classification applications. Data range and Intensity, (strength of the backscattered signals measured by the LiDAR systems), are affected by the flying height, the ground elevation, scanning angle and the physical characteristics of the objects surface. These effects may lead to uneven distribution of point cloud or some gaps that may affect the classification process. Researchers have investigated the conversion of LiDAR range point data to raster image for terrain modelling. Interpolation techniques have been used to achieve the best representation of surfaces, and to fill the gaps between the LiDAR footprints. Interpolation methods are also investigated to generate LiDAR range and intensity image data for land cover classification applications. In this paper, different approach has been followed to classifying the LiDAR data (range and intensity) for land cover mapping. The methodology relies on the classification of the point cloud data based on their range and intensity and then converted the classified points into raster image. The gaps in the data are filled based on the classes of the nearest neighbour. Land cover maps are produced using two approaches using: (a) the conventional raster image data based on point interpolation; and (b) the proposed point data classification. A study area covering an urban district in Burnaby, British Colombia, Canada, is selected to compare the results of the two approaches. Five different land cover classes can be distinguished in that area: buildings, roads and parking areas, trees, low vegetation (grass), and bare soil. The results show that an improvement of around 10 % in the
Errors, error detection, error correction and hippocampal-region damage: data and theories.
MacKay, Donald G; Johnson, Laura W
2013-11-01
This review and perspective article outlines 15 observational constraints on theories of errors, error detection, and error correction, and their relation to hippocampal-region (HR) damage. The core observations come from 10 studies with H.M., an amnesic with cerebellar and HR damage but virtually no neocortical damage. Three studies examined the detection of errors planted in visual scenes (e.g., a bird flying in a fish bowl in a school classroom) and sentences (e.g., I helped themselves to the birthday cake). In all three experiments, H.M. detected reliably fewer errors than carefully matched memory-normal controls. Other studies examined the detection and correction of self-produced errors, with controls for comprehension of the instructions, impaired visual acuity, temporal factors, motoric slowing, forgetting, excessive memory load, lack of motivation, and deficits in visual scanning or attention. In these studies, H.M. corrected reliably fewer errors than memory-normal and cerebellar controls, and his uncorrected errors in speech, object naming, and reading aloud exhibited two consistent features: omission and anomaly. For example, in sentence production tasks, H.M. omitted one or more words in uncorrected encoding errors that rendered his sentences anomalous (incoherent, incomplete, or ungrammatical) reliably more often than controls. Besides explaining these core findings, the theoretical principles discussed here explain H.M.'s retrograde amnesia for once familiar episodic and semantic information; his anterograde amnesia for novel information; his deficits in visual cognition, sentence comprehension, sentence production, sentence reading, and object naming; and effects of aging on his ability to read isolated low frequency words aloud. These theoretical principles also explain a wide range of other data on error detection and correction and generate new predictions for future test. Copyright © 2013 Elsevier Ltd. All rights reserved.
Discriminative Nonlinear Analysis Operator Learning: When Cosparse Model Meets Image Classification.
Wen, Zaidao; Hou, Biao; Jiao, Licheng
2017-05-03
Linear synthesis model based dictionary learning framework has achieved remarkable performances in image classification in the last decade. Behaved as a generative feature model, it however suffers from some intrinsic deficiencies. In this paper, we propose a novel parametric nonlinear analysis cosparse model (NACM) with which a unique feature vector will be much more efficiently extracted. Additionally, we derive a deep insight to demonstrate that NACM is capable of simultaneously learning the task adapted feature transformation and regularization to encode our preferences, domain prior knowledge and task oriented supervised information into the features. The proposed NACM is devoted to the classification task as a discriminative feature model and yield a novel discriminative nonlinear analysis operator learning framework (DNAOL). The theoretical analysis and experimental performances clearly demonstrate that DNAOL will not only achieve the better or at least competitive classification accuracies than the state-of-the-art algorithms but it can also dramatically reduce the time complexities in both training and testing phases.
Minimum-error discrimination of entangled quantum states
International Nuclear Information System (INIS)
Lu, Y.; Coish, N.; Kaltenbaek, R.; Hamel, D. R.; Resch, K. J.; Croke, S.
2010-01-01
Strategies to optimally discriminate between quantum states are critical in quantum technologies. We present an experimental demonstration of minimum-error discrimination between entangled states, encoded in the polarization of pairs of photons. Although the optimal measurement involves projection onto entangled states, we use a result of J. Walgate et al. [Phys. Rev. Lett. 85, 4972 (2000)] to design an optical implementation employing only local polarization measurements and feed-forward, which performs at the Helstrom bound. Our scheme can achieve perfect discrimination of orthogonal states and minimum-error discrimination of nonorthogonal states. Our experimental results show a definite advantage over schemes not using feed-forward.
Teffahi, Hanane; Yao, Hongxun; Belabid, Nasreddine; Chaib, Souleyman
2018-02-01
The satellite images with very high spatial resolution have been recently widely used in image classification topic as it has become challenging task in remote sensing field. Due to a number of limitations such as the redundancy of features and the high dimensionality of the data, different classification methods have been proposed for remote sensing images classification particularly the methods using feature extraction techniques. This paper propose a simple efficient method exploiting the capability of extended multi-attribute profiles (EMAP) with sparse autoencoder (SAE) for remote sensing image classification. The proposed method is used to classify various remote sensing datasets including hyperspectral and multispectral images by extracting spatial and spectral features based on the combination of EMAP and SAE by linking them to kernel support vector machine (SVM) for classification. Experiments on new hyperspectral image "Huston data" and multispectral image "Washington DC data" shows that this new scheme can achieve better performance of feature learning than the primitive features, traditional classifiers and ordinary autoencoder and has huge potential to achieve higher accuracy for classification in short running time.
Hyperspectral Image Classification Using Discriminative Dictionary Learning
International Nuclear Information System (INIS)
Zongze, Y; Hao, S; Kefeng, J; Huanxin, Z
2014-01-01
The hyperspectral image (HSI) processing community has witnessed a surge of papers focusing on the utilization of sparse prior for effective HSI classification. In sparse representation based HSI classification, there are two phases: sparse coding with an over-complete dictionary and classification. In this paper, we first apply a novel fisher discriminative dictionary learning method, which capture the relative difference in different classes. The competitive selection strategy ensures that atoms in the resulting over-complete dictionary are the most discriminative. Secondly, motivated by the assumption that spatially adjacent samples are statistically related and even belong to the same materials (same class), we propose a majority voting scheme incorporating contextual information to predict the category label. Experiment results show that the proposed method can effectively strengthen relative discrimination of the constructed dictionary, and incorporating with the majority voting scheme achieve generally an improved prediction performance
An ordinal classification approach for CTG categorization.
Georgoulas, George; Karvelis, Petros; Gavrilis, Dimitris; Stylios, Chrysostomos D; Nikolakopoulos, George
2017-07-01
Evaluation of cardiotocogram (CTG) is a standard approach employed during pregnancy and delivery. But, its interpretation requires high level expertise to decide whether the recording is Normal, Suspicious or Pathological. Therefore, a number of attempts have been carried out over the past three decades for development automated sophisticated systems. These systems are usually (multiclass) classification systems that assign a category to the respective CTG. However most of these systems usually do not take into consideration the natural ordering of the categories associated with CTG recordings. In this work, an algorithm that explicitly takes into consideration the ordering of CTG categories, based on binary decomposition method, is investigated. Achieved results, using as a base classifier the C4.5 decision tree classifier, prove that the ordinal classification approach is marginally better than the traditional multiclass classification approach, which utilizes the standard C4.5 algorithm for several performance criteria.
Remote one-qubit information concentration and decoding of operator quantum error-correction codes
International Nuclear Information System (INIS)
Hsu Liyi
2007-01-01
We propose the general scheme of remote one-qubit information concentration. To achieve the task, the Bell-correlated mixed states are exploited. In addition, the nonremote one-qubit information concentration is equivalent to the decoding of the quantum error-correction code. Here we propose how to decode the stabilizer codes. In particular, the proposed scheme can be used for the operator quantum error-correction codes. The encoded state can be recreated on the errorless qubit, regardless how many bit-flip errors and phase-flip errors have occurred
Olives, Casey; Valadez, Joseph J; Brooker, Simon J; Pagano, Marcello
2012-01-01
Originally a binary classifier, Lot Quality Assurance Sampling (LQAS) has proven to be a useful tool for classification of the prevalence of Schistosoma mansoni into multiple categories (≤10%, >10 and LQAS (MC-LQAS) have not received full treatment. We explore the analytical properties of MC-LQAS, and validate its use for the classification of S. mansoni prevalence in multiple settings in East Africa. We outline MC-LQAS design principles and formulae for operating characteristic curves. In addition, we derive the average sample number for MC-LQAS when utilizing semi-curtailed sampling and introduce curtailed sampling in this setting. We also assess the performance of MC-LQAS designs with maximum sample sizes of n=15 and n=25 via a weighted kappa-statistic using S. mansoni data collected in 388 schools from four studies in East Africa. Overall performance of MC-LQAS classification was high (kappa-statistic of 0.87). In three of the studies, the kappa-statistic for a design with n=15 was greater than 0.75. In the fourth study, where these designs performed poorly (kappa-statistic less than 0.50), the majority of observations fell in regions where potential error is known to be high. Employment of semi-curtailed and curtailed sampling further reduced the sample size by as many as 0.5 and 3.5 observations per school, respectively, without increasing classification error. This work provides the needed analytics to understand the properties of MC-LQAS for assessing the prevalance of S. mansoni and shows that in most settings a sample size of 15 children provides a reliable classification of schools.
[Diagnosis and classification of headache and temporomandibular disorders, a new opportunity].
Koole, P; Koole, R
2011-05-01
Patients with orofacial pains are generally treated by physicians. A small number of patients are treated for pain in the temporomandibular joint, the masticatory and the neck muscles, by dentists and orofacial surgeons. Among half of the patients being treated in neurological headache clinics, the temporomandibular joint and the masticatory muscles are the source of the pain. In order to achieve better research and a classification, the International Headache Society, consisting largely of neurologists, developed a classification system. A comparable development occurred among oral health specialists. Employing these 2 methods with the same patients leads to different diagnoses and treatments. Both the International Classification of Headache Disorders II and the Research Diagnostic Criteria for Temporomandibular Disorders are being revised. This creates the opportunity to establish a single classification for these orofacial pains, preferably within the new International Classification of Headache Disorders.
Coherent error study in a retarding field energy analyzer
International Nuclear Information System (INIS)
Cui, Y.; Zou, Y.; Reiser, M.; Kishek, R.A.; Haber, I.; Bernal, S.; O'Shea, P.G.
2005-01-01
A novel cylindrical retarding electrostatic field energy analyzer for low-energy beams has been designed, simulated, and tested with electron beams of several keV, in which space charge effects play an important role. A cylindrical focusing electrode is used to overcome the beam expansion inside the device due to space-charge forces, beam emittance, etc. In this paper, we present the coherent error analysis for this energy analyzer with beam envelope equation including space charge and emittance effects. The study shows that this energy analyzer can achieve very high resolution (with relative error of around 10 -5 ) if taking away the coherent errors by using proper focusing voltages. The theoretical analysis is compared with experimental results
Capacitor Mismatch Error Cancellation Technique for a Successive Approximation A/D Converter
DEFF Research Database (Denmark)
Zheng, Zhiliang; Moon, Un-Ku; Steensgaard-Madsen, Jesper
1999-01-01
An error cancellation technique is described for suppressing capacitor mismatch in a successive approximation A/D converter. At the cost of a 50% increase in conversion time, the first-order capacitor mismatch error is cancelled. Methods for achieving top-plate parasitic insensitive operation...... are described, and the use of a gain- and offset-compensated opamp is explained. SWITCAP simulation results show that the proposed 16-bit SAR ADC can achieve an SNDR of over 91 dB under non-ideal conditions, including 1% 3 sigma nominal capacitor mismatch, 10-20% randomized parasitic capacitors, 66 dB opamp...
Feature generation and representations for protein-protein interaction classification.
Lan, Man; Tan, Chew Lim; Su, Jian
2009-10-01
Automatic detecting protein-protein interaction (PPI) relevant articles is a crucial step for large-scale biological database curation. The previous work adopted POS tagging, shallow parsing and sentence splitting techniques, but they achieved worse performance than the simple bag-of-words representation. In this paper, we generated and investigated multiple types of feature representations in order to further improve the performance of PPI text classification task. Besides the traditional domain-independent bag-of-words approach and the term weighting methods, we also explored other domain-dependent features, i.e. protein-protein interaction trigger keywords, protein named entities and the advanced ways of incorporating Natural Language Processing (NLP) output. The integration of these multiple features has been evaluated on the BioCreAtIvE II corpus. The experimental results showed that both the advanced way of using NLP output and the integration of bag-of-words and NLP output improved the performance of text classification. Specifically, in comparison with the best performance achieved in the BioCreAtIvE II IAS, the feature-level and classifier-level integration of multiple features improved the performance of classification 2.71% and 3.95%, respectively.
LDA boost classification: boosting by topics
Lei, La; Qiao, Guo; Qimin, Cao; Qitao, Li
2012-12-01
AdaBoost is an efficacious classification algorithm especially in text categorization (TC) tasks. The methodology of setting up a classifier committee and voting on the documents for classification can achieve high categorization precision. However, traditional Vector Space Model can easily lead to the curse of dimensionality and feature sparsity problems; so it affects classification performance seriously. This article proposed a novel classification algorithm called LDABoost based on boosting ideology which uses Latent Dirichlet Allocation (LDA) to modeling the feature space. Instead of using words or phrase, LDABoost use latent topics as the features. In this way, the feature dimension is significantly reduced. Improved Naïve Bayes (NB) is designed as the weaker classifier which keeps the efficiency advantage of classic NB algorithm and has higher precision. Moreover, a two-stage iterative weighted method called Cute Integration in this article is proposed for improving the accuracy by integrating weak classifiers into strong classifier in a more rational way. Mutual Information is used as metrics of weights allocation. The voting information and the categorization decision made by basis classifiers are fully utilized for generating the strong classifier. Experimental results reveals LDABoost making categorization in a low-dimensional space, it has higher accuracy than traditional AdaBoost algorithms and many other classic classification algorithms. Moreover, its runtime consumption is lower than different versions of AdaBoost, TC algorithms based on support vector machine and Neural Networks.
Reader error during CT colonography: causes and implications for training
International Nuclear Information System (INIS)
Slater, Andrew; Tam, Emily; Gartner, Louise; Scarth, Julia; Peiris, Chand; Gupta, Arun; Marshall, Michele; Burling, David; Taylor, Stuart A.; Halligan, Steve
2006-01-01
This study investigated the variability in baseline computed tomography colonography (CTC) performance using untrained readers by documenting sources of error to guide future training requirements. Twenty CTC endoscopically validated data sets containing 32 polyps were consensus read by three unblinded radiologists experienced in CTC, creating a reference standard. Six readers without prior CTC training [four residents and two board-certified subspecialty gastrointestinal (GI) radiologists] read the 20 cases. Readers drew a region of interest (ROI) around every area they considered a potential colonic lesion, even if subsequently dismissed, before creating a final report. Using this final report, reader ROIs were classified as true positive detections, true negatives correctly dismissed, true detections incorrectly dismissed (i.e., classification error), or perceptual errors. Detection of polyps 1-5 mm, 6-9 mm, and ≥10 mm ranged from 7.1% to 28.6%, 16.7% to 41.7%, and 16.7% to 83.3%, respectively. There was no significant difference between polyp detection or false positives for the GI radiologists compared with residents (p=0.67, p=0.4 respectively). Most missed polyps were due to failure of detection rather than characterization (range 82-95%). Untrained reader performance is variable but generally poor. Most missed polyps are due perceptual error rather than characterization, suggesting basic training should focus heavily on lesion detection. (orig.)
Hyvärinen, A
1985-01-01
The main purpose of the present study was to describe the statistical behaviour of daily analytical errors in the dimensions of place and time, providing a statistical basis for realistic estimates of the analytical error, and hence allowing the importance of the error and the relative contributions of its different sources to be re-evaluated. The observation material consists of creatinine and glucose results for control sera measured in daily routine quality control in five laboratories for a period of one year. The observation data were processed and computed by means of an automated data processing system. Graphic representations of time series of daily observations, as well as their means and dispersion limits when grouped over various time intervals, were investigated. For partition of the total variation several two-way analyses of variance were done with laboratory and various time classifications as factors. Pooled sets of observations were tested for normality of distribution and for consistency of variances, and the distribution characteristics of error variation in different categories of place and time were compared. Errors were found from the time series to vary typically between days. Due to irregular fluctuations in general and particular seasonal effects in creatinine, stable estimates of means or of dispersions for errors in individual laboratories could not be easily obtained over short periods of time but only from data sets pooled over long intervals (preferably at least one year). Pooled estimates of proportions of intralaboratory variation were relatively low (less than 33%) when the variation was pooled within days. However, when the variation was pooled over longer intervals this proportion increased considerably, even to a maximum of 89-98% (95-98% in each method category) when an outlying laboratory in glucose was omitted, with a concomitant decrease in the interaction component (representing laboratory-dependent variation with time
Wu, Jie; Besnehard, Quentin; Marchessoux, Cédric
2011-03-01
Clinical studies for the validation of new medical imaging devices require hundreds of images. An important step in creating and tuning the study protocol is the classification of images into "difficult" and "easy" cases. This consists of classifying the image based on features like the complexity of the background, the visibility of the disease (lesions). Therefore, an automatic medical background classification tool for mammograms would help for such clinical studies. This classification tool is based on a multi-content analysis framework (MCA) which was firstly developed to recognize image content of computer screen shots. With the implementation of new texture features and a defined breast density scale, the MCA framework is able to automatically classify digital mammograms with a satisfying accuracy. BI-RADS (Breast Imaging Reporting Data System) density scale is used for grouping the mammograms, which standardizes the mammography reporting terminology and assessment and recommendation categories. Selected features are input into a decision tree classification scheme in MCA framework, which is the so called "weak classifier" (any classifier with a global error rate below 50%). With the AdaBoost iteration algorithm, these "weak classifiers" are combined into a "strong classifier" (a classifier with a low global error rate) for classifying one category. The results of classification for one "strong classifier" show the good accuracy with the high true positive rates. For the four categories the results are: TP=90.38%, TN=67.88%, FP=32.12% and FN =9.62%.
Kakkos, I.; Gkiatis, K.; Bromis, K.; Asvestas, P. A.; Karanasiou, I. S.; Ventouras, E. M.; Matsopoulos, G. K.
2017-11-01
The detection of an error is the cognitive evaluation of an action outcome that is considered undesired or mismatches an expected response. Brain activity during monitoring of correct and incorrect responses elicits Event Related Potentials (ERPs) revealing complex cerebral responses to deviant sensory stimuli. Development of accurate error detection systems is of great importance both concerning practical applications and in investigating the complex neural mechanisms of decision making. In this study, data are used from an audio identification experiment that was implemented with two levels of complexity in order to investigate neurophysiological error processing mechanisms in actors and observers. To examine and analyse the variations of the processing of erroneous sensory information for each level of complexity we employ Support Vector Machines (SVM) classifiers with various learning methods and kernels using characteristic ERP time-windowed features. For dimensionality reduction and to remove redundant features we implement a feature selection framework based on Sequential Forward Selection (SFS). The proposed method provided high accuracy in identifying correct and incorrect responses both for actors and for observers with mean accuracy of 93% and 91% respectively. Additionally, computational time was reduced and the effects of the nesting problem usually occurring in SFS of large feature sets were alleviated.
Reproducibility of the Lauge-Hansen, Danis-Weber, and AO classifications for ankle fractures
Directory of Open Access Journals (Sweden)
Lucas Lopes da Fonseca
Full Text Available ABSTRACT Objective: This study evaluated the reproducibility of the three main classifications of ankle fractures most commonly used in emergency clinical practice: Lauge-Hansen, Danis-Weber, and AO-OTA. The secondary objective was to assess whether the level of professional experience influenced the interobserver agreement for the classification of this pathology. Methods: The study included 83 digitized preoperative radiographic images of ankle fractures, in anteroposterior and lateral views, of different adults that had occurred between January and December 2013. For sample calculation, the estimated accuracy was approximately 15%, with a sampling error of 5% and a sampling power of 80%. The images were analyzed and classified by six different observers: two foot and ankle surgeons, two general orthopedic surgeons, and two-second-year residents in orthopedics and traumatology. The Kappa statistical method of multiple variances was used to assess the variations. Results: The Danis-Weber classification indicated that 40% of the agreements among all observers were good or excellent, whereas only 20% of good and excellent agreements were obtained using the AO and Lauge Hansen classifications. The Kappa index was 0.49 for the Danis-Weber classification, 0.32 for Lauge Hansen, and 0.38 for AO. Conclusion: The Hansen-Lauge classification presented the poorest interobserver agreement among the three systems. The AO classification demonstrated a moderate agreement and the Danis-Weber classification presented an excellent interobserver agreement index, regardless of professional experience.
Groundwater level prediction of landslide based on classification and regression tree
Directory of Open Access Journals (Sweden)
Yannan Zhao
2016-09-01
Full Text Available According to groundwater level monitoring data of Shuping landslide in the Three Gorges Reservoir area, based on the response relationship between influential factors such as rainfall and reservoir level and the change of groundwater level, the influential factors of groundwater level were selected. Then the classification and regression tree (CART model was constructed by the subset and used to predict the groundwater level. Through the verification, the predictive results of the test sample were consistent with the actually measured values, and the mean absolute error and relative error is 0.28 m and 1.15% respectively. To compare the support vector machine (SVM model constructed using the same set of factors, the mean absolute error and relative error of predicted results is 1.53 m and 6.11% respectively. It is indicated that CART model has not only better fitting and generalization ability, but also strong advantages in the analysis of landslide groundwater dynamic characteristics and the screening of important variables. It is an effective method for prediction of ground water level in landslides.
A Classification-based Review Recommender
O'Mahony, Michael P.; Smyth, Barry
Many online stores encourage their users to submit product/service reviews in order to guide future purchasing decisions. These reviews are often listed alongside product recommendations but, to date, limited attention has been paid as to how best to present these reviews to the end-user. In this paper, we describe a supervised classification approach that is designed to identify and recommend the most helpful product reviews. Using the TripAdvisor service as a case study, we compare the performance of several classification techniques using a range of features derived from hotel reviews. We then describe how these classifiers can be used as the basis for a practical recommender that automatically suggests the mosthelpful contrasting reviews to end-users. We present an empirical evaluation which shows that our approach achieves a statistically significant improvement over alternative review ranking schemes.
Huo, Ming-Xia; Li, Ying
2017-12-01
Quantum error correction is important to quantum information processing, which allows us to reliably process information encoded in quantum error correction codes. Efficient quantum error correction benefits from the knowledge of error rates. We propose a protocol for monitoring error rates in real time without interrupting the quantum error correction. Any adaptation of the quantum error correction code or its implementation circuit is not required. The protocol can be directly applied to the most advanced quantum error correction techniques, e.g. surface code. A Gaussian processes algorithm is used to estimate and predict error rates based on error correction data in the past. We find that using these estimated error rates, the probability of error correction failures can be significantly reduced by a factor increasing with the code distance.
Controlling qubit drift by recycling error correction syndromes
Blume-Kohout, Robin
2015-03-01
Physical qubits are susceptible to systematic drift, above and beyond the stochastic Markovian noise that motivates quantum error correction. This parameter drift must be compensated - if it is ignored, error rates will rise to intolerable levels - but compensation requires knowing the parameters' current value, which appears to require halting experimental work to recalibrate (e.g. via quantum tomography). Fortunately, this is untrue. I show how to perform on-the-fly recalibration on the physical qubits in an error correcting code, using only information from the error correction syndromes. The algorithm for detecting and compensating drift is very simple - yet, remarkably, when used to compensate Brownian drift in the qubit Hamiltonian, it achieves a stabilized error rate very close to the theoretical lower bound. Against 1/f noise, it is less effective only because 1/f noise is (like white noise) dominated by high-frequency fluctuations that are uncompensatable. Sandia National Laboratories is a multi-program laboratory managed and operated by Sandia Corporation, a wholly owned subsidiary of Lockheed Martin Corporation, for the U.S. Department of Energy's National Nuclear Security Administration under contract DE
THERP and HEART integrated methodology for human error assessment
Castiglia, Francesco; Giardina, Mariarosa; Tomarchio, Elio
2015-11-01
THERP and HEART integrated methodology is proposed to investigate accident scenarios that involve operator errors during high-dose-rate (HDR) treatments. The new approach has been modified on the basis of fuzzy set concept with the aim of prioritizing an exhaustive list of erroneous tasks that can lead to patient radiological overexposures. The results allow for the identification of human errors that are necessary to achieve a better understanding of health hazards in the radiotherapy treatment process, so that it can be properly monitored and appropriately managed.
Classification of breast cancer cytological specimen using convolutional neural network
Żejmo, Michał; Kowal, Marek; Korbicz, Józef; Monczak, Roman
2017-01-01
The paper presents a deep learning approach for automatic classification of breast tumors based on fine needle cytology. The main aim of the system is to distinguish benign from malignant cases based on microscopic images. Experiment was carried out on cytological samples derived from 50 patients (25 benign cases + 25 malignant cases) diagnosed in Regional Hospital in Zielona Góra. To classify microscopic images, we used convolutional neural networks (CNN) of two types: GoogLeNet and AlexNet. Due to the very large size of images of cytological specimen (on average 200000 × 100000 pixels), they were divided into smaller patches of size 256 × 256 pixels. Breast cancer classification usually is based on morphometric features of nuclei. Therefore, training and validation patches were selected using Support Vector Machine (SVM) so that suitable amount of cell material was depicted. Neural classifiers were tuned using GPU accelerated implementation of gradient descent algorithm. Training error was defined as a cross-entropy classification loss. Classification accuracy was defined as the percentage ratio of successfully classified validation patches to the total number of validation patches. The best accuracy rate of 83% was obtained by GoogLeNet model. We observed that more misclassified patches belong to malignant cases.
Gaia eclipsing binary and multiple systems. Supervised classification and self-organizing maps
Süveges, M.; Barblan, F.; Lecoeur-Taïbi, I.; Prša, A.; Holl, B.; Eyer, L.; Kochoska, A.; Mowlavi, N.; Rimoldini, L.
2017-07-01
Context. Large surveys producing tera- and petabyte-scale databases require machine-learning and knowledge discovery methods to deal with the overwhelming quantity of data and the difficulties of extracting concise, meaningful information with reliable assessment of its uncertainty. This study investigates the potential of a few machine-learning methods for the automated analysis of eclipsing binaries in the data of such surveys. Aims: We aim to aid the extraction of samples of eclipsing binaries from such databases and to provide basic information about the objects. We intend to estimate class labels according to two different, well-known classification systems, one based on the light curve morphology (EA/EB/EW classes) and the other based on the physical characteristics of the binary system (system morphology classes; detached through overcontact systems). Furthermore, we explore low-dimensional surfaces along which the light curves of eclipsing binaries are concentrated, and consider their use in the characterization of the binary systems and in the exploration of biases of the full unknown Gaia data with respect to the training sets. Methods: We have explored the performance of principal component analysis (PCA), linear discriminant analysis (LDA), Random Forest classification and self-organizing maps (SOM) for the above aims. We pre-processed the photometric time series by combining a double Gaussian profile fit and a constrained smoothing spline, in order to de-noise and interpolate the observed light curves. We achieved further denoising, and selected the most important variability elements from the light curves using PCA. Supervised classification was performed using Random Forest and LDA based on the PC decomposition, while SOM gives a continuous 2-dimensional manifold of the light curves arranged by a few important features. We estimated the uncertainty of the supervised methods due to the specific finite training set using ensembles of models constructed
Toward noncooperative iris recognition: a classification approach using multiple signatures.
Proença, Hugo; Alexandre, Luís A
2007-04-01
This paper focuses on noncooperative iris recognition, i.e., the capture of iris images at large distances, under less controlled lighting conditions, and without active participation of the subjects. This increases the probability of capturing very heterogeneous images (regarding focus, contrast, or brightness) and with several noise factors (iris obstructions and reflections). Current iris recognition systems are unable to deal with noisy data and substantially increase their error rates, especially the false rejections, in these conditions. We propose an iris classification method that divides the segmented and normalized iris image into six regions, makes an independent feature extraction and comparison for each region, and combines each of the dissimilarity values through a classification rule. Experiments show a substantial decrease, higher than 40 percent, of the false rejection rates in the recognition of noisy iris images.
On the Design of Error-Correcting Ciphers
Directory of Open Access Journals (Sweden)
Mathur Chetan Nanjunda
2006-01-01
Full Text Available Securing transmission over a wireless network is especially challenging, not only because of the inherently insecure nature of the medium, but also because of the highly error-prone nature of the wireless environment. In this paper, we take a joint encryption-error correction approach to ensure secure and robust communication over the wireless link. In particular, we design an error-correcting cipher (called the high diffusion cipher and prove bounds on its error-correcting capacity as well as its security. Towards this end, we propose a new class of error-correcting codes (HD-codes with built-in security features that we use in the diffusion layer of the proposed cipher. We construct an example, 128-bit cipher using the HD-codes, and compare it experimentally with two traditional concatenated systems: (a AES (Rijndael followed by Reed-Solomon codes, (b Rijndael followed by convolutional codes. We show that the HD-cipher is as resistant to linear and differential cryptanalysis as the Rijndael. We also show that any chosen plaintext attack that can be performed on the HD cipher can be transformed into a chosen plaintext attack on the Rijndael cipher. In terms of error correction capacity, the traditional systems using Reed-Solomon codes are comparable to the proposed joint error-correcting cipher and those that use convolutional codes require more data expansion in order to achieve similar error correction as the HD-cipher. The original contributions of this work are (1 design of a new joint error-correction-encryption system, (2 design of a new class of algebraic codes with built-in security criteria, called the high diffusion codes (HD-codes for use in the HD-cipher, (3 mathematical properties of these codes, (4 methods for construction of the codes, (5 bounds on the error-correcting capacity of the HD-cipher, (6 mathematical derivation of the bound on resistance of HD cipher to linear and differential cryptanalysis, (7 experimental comparison
Directory of Open Access Journals (Sweden)
H. Y. Gu
2017-09-01
Full Text Available Classification rule set is important for Land Cover classification, which refers to features and decision rules. The selection of features and decision are based on an iterative trial-and-error approach that is often utilized in GEOBIA, however, it is time-consuming and has a poor versatility. This study has put forward a rule set building method for Land cover classification based on human knowledge and machine learning. The use of machine learning is to build rule sets effectively which will overcome the iterative trial-and-error approach. The use of human knowledge is to solve the shortcomings of existing machine learning method on insufficient usage of prior knowledge, and improve the versatility of rule sets. A two-step workflow has been introduced, firstly, an initial rule is built based on Random Forest and CART decision tree. Secondly, the initial rule is analyzed and validated based on human knowledge, where we use statistical confidence interval to determine its threshold. The test site is located in Potsdam City. We utilised the TOP, DSM and ground truth data. The results show that the method could determine rule set for Land Cover classification semi-automatically, and there are static features for different land cover classes.
Medication errors: prescribing faults and prescription errors.
Velo, Giampaolo P; Minuz, Pietro
2009-06-01
1. Medication errors are common in general practice and in hospitals. Both errors in the act of writing (prescription errors) and prescribing faults due to erroneous medical decisions can result in harm to patients. 2. Any step in the prescribing process can generate errors. Slips, lapses, or mistakes are sources of errors, as in unintended omissions in the transcription of drugs. Faults in dose selection, omitted transcription, and poor handwriting are common. 3. Inadequate knowledge or competence and incomplete information about clinical characteristics and previous treatment of individual patients can result in prescribing faults, including the use of potentially inappropriate medications. 4. An unsafe working environment, complex or undefined procedures, and inadequate communication among health-care personnel, particularly between doctors and nurses, have been identified as important underlying factors that contribute to prescription errors and prescribing faults. 5. Active interventions aimed at reducing prescription errors and prescribing faults are strongly recommended. These should be focused on the education and training of prescribers and the use of on-line aids. The complexity of the prescribing procedure should be reduced by introducing automated systems or uniform prescribing charts, in order to avoid transcription and omission errors. Feedback control systems and immediate review of prescriptions, which can be performed with the assistance of a hospital pharmacist, are also helpful. Audits should be performed periodically.
Learning a locomotor task: with or without errors?
Marchal-Crespo, Laura; Schneider, Jasmin; Jaeger, Lukas; Riener, Robert
2014-03-04
Robotic haptic guidance is the most commonly used robotic training strategy to reduce performance errors while training. However, research on motor learning has emphasized that errors are a fundamental neural signal that drive motor adaptation. Thus, researchers have proposed robotic therapy algorithms that amplify movement errors rather than decrease them. However, to date, no study has analyzed with precision which training strategy is the most appropriate to learn an especially simple task. In this study, the impact of robotic training strategies that amplify or reduce errors on muscle activation and motor learning of a simple locomotor task was investigated in twenty two healthy subjects. The experiment was conducted with the MAgnetic Resonance COmpatible Stepper (MARCOS) a special robotic device developed for investigations in the MR scanner. The robot moved the dominant leg passively and the subject was requested to actively synchronize the non-dominant leg to achieve an alternating stepping-like movement. Learning with four different training strategies that reduce or amplify errors was evaluated: (i) Haptic guidance: errors were eliminated by passively moving the limbs, (ii) No guidance: no robot disturbances were presented, (iii) Error amplification: existing errors were amplified with repulsive forces, (iv) Noise disturbance: errors were evoked intentionally with a randomly-varying force disturbance on top of the no guidance strategy. Additionally, the activation of four lower limb muscles was measured by the means of surface electromyography (EMG). Strategies that reduce or do not amplify errors limit muscle activation during training and result in poor learning gains. Adding random disturbing forces during training seems to increase attention, and therefore improve motor learning. Error amplification seems to be the most suitable strategy for initially less skilled subjects, perhaps because subjects could better detect their errors and correct them
FIASCO II failure to achieve a satisfactory cardiac outcome study: the elimination of system errors.
Farid, Shakil; Page, Aravinda; Jenkins, David; Jones, Mark T; Freed, Darren; Nashef, Samer A M
2013-07-01
Death in low-risk cardiac surgical patients provides a simple and accessible method by which modifiable causes of death can be identified. In the first FIASCO study published in 2009, local potentially modifiable causes of preventable death in low-risk patients with a logistic EuroSCORE of 0-2 undergoing cardiac surgery were inadequate myocardial protection and lack of clarity in the chain of responsibility. As a result, myocardial protection was improved, and a formalized system introduced to ensure clarity of the chain of responsibility in the care of all cardiac surgical patients. The purpose of the current study was to re-audit outcomes in low-risk patients to see if improvements have been achieved. Patients with a logistic EuroSCORE of 0-2 who had cardiac surgery from January 2006 to August 2012 were included. Data were prospectively collected and retrospectively analysed. The case notes of patients who died in hospital were subject to internal and external review and classified according to preventability. Two thousand five hundred and forty-nine patients with a logistic EuroSCORE of 0-2 underwent cardiac surgery during the study period. Seven deaths occurred in truly low-risk patients, giving a mortality of 0.27%. Of the seven, three were considered preventable and four non-preventable. Mortality was marginally lower than in our previous study (0.37%), and no death occurred as a result of inadequate myocardial protection or communication failures. We postulate that the regular study of such events in all institutions may unmask systemic errors that can be remedied to prevent or reduce future occurrences. We encourage all units to use this methodology to detect any similarly modifiable factors in their practice.
Catchment Classification: Connecting Climate, Structure and Function
Sawicz, K. A.; Wagener, T.; Sivapalan, M.; Troch, P. A.; Carrillo, G. A.
2010-12-01
Hydrology does not yet possess a generally accepted catchment classification framework. Such a classification framework needs to: [1] give names to things, i.e. the main classification step, [2] permit transfer of information, i.e. regionalization of information, [3] permit development of generalizations, i.e. to develop new theory, and [4] provide a first order environmental change impact assessment, i.e., the hydrologic implications of climate, land use and land cover change. One strategy is to create a catchment classification framework based on the notion of catchment functions (partitioning, storage, and release). Results of an empirical study presented here connects climate and structure to catchment function (in the form of select hydrologic signatures), based on analyzing over 300 US catchments. Initial results indicate a wide assortment of signature relationships with properties of climate, geology, and vegetation. The uncertainty in the different regionalized signatures varies widely, and therefore there is variability in the robustness of classifying ungauged basins. This research provides insight into the controls of hydrologic behavior of a catchment, and enables a classification framework applicable to gauged and ungauged across the study domain. This study sheds light on what we can expect to achieve in mapping climate, structure and function in a top-down manner. Results of this study complement work done using a bottom-up physically-based modeling framework to generalize this approach (Carrillo et al., this session).
Wan, Xiaoqing; Zhao, Chunhui
2017-06-01
As a competitive machine learning algorithm, the stacked sparse autoencoder (SSA) has achieved outstanding popularity in exploiting high-level features for classification of hyperspectral images (HSIs). In general, in the SSA architecture, the nodes between adjacent layers are fully connected and need to be iteratively fine-tuned during the pretraining stage; however, the nodes of previous layers further away may be less likely to have a dense correlation to the given node of subsequent layers. Therefore, to reduce the classification error and increase the learning rate, this paper proposes the general framework of locally connected SSA; that is, the biologically inspired local receptive field (LRF) constrained SSA architecture is employed to simultaneously characterize the local correlations of spectral features and extract high-level feature representations of hyperspectral data. In addition, the appropriate receptive field constraint is concurrently updated by measuring the spatial distances from the neighbor nodes to the corresponding node. Finally, the efficient random forest classifier is cascaded to the last hidden layer of the SSA architecture as a benchmark classifier. Experimental results on two real HSI datasets demonstrate that the proposed hierarchical LRF constrained stacked sparse autoencoder and random forest (SSARF) provides encouraging results with respect to other contrastive methods, for instance, the improvements of overall accuracy in a range of 0.72%-10.87% for the Indian Pines dataset and 0.74%-7.90% for the Kennedy Space Center dataset; moreover, it generates lower running time compared with the result provided by similar SSARF based methodology.
Using ecological zones to increase the detail of Landsat classifications
Fox, L., III; Mayer, K. E.
1981-01-01
Changes in classification detail of forest species descriptions were made for Landsat data on 2.2 million acres in northwestern California. Because basic forest canopy structures may exhibit very similar E-M energy reflectance patterns in different environmental regions, classification labels based on Landsat spectral signatures alone become very generalized when mapping large heterogeneous ecological regions. By adding a seven ecological zone stratification, a 167% improvement in classification detail was made over the results achieved without it. The seven zone stratification is a less costly alternative to the inclusion of complex collateral information, such as terrain data and soil type, into the Landsat data base when making inventories of areas greater than 500,000 acres.
Land use/cover classification in the Brazilian Amazon using satellite images.
Lu, Dengsheng; Batistella, Mateus; Li, Guiying; Moran, Emilio; Hetrick, Scott; Freitas, Corina da Costa; Dutra, Luciano Vieira; Sant'anna, Sidnei João Siqueira
2012-09-01
Land use/cover classification is one of the most important applications in remote sensing. However, mapping accurate land use/cover spatial distribution is a challenge, particularly in moist tropical regions, due to the complex biophysical environment and limitations of remote sensing data per se. This paper reviews experiments related to land use/cover classification in the Brazilian Amazon for a decade. Through comprehensive analysis of the classification results, it is concluded that spatial information inherent in remote sensing data plays an essential role in improving land use/cover classification. Incorporation of suitable textural images into multispectral bands and use of segmentation-based method are valuable ways to improve land use/cover classification, especially for high spatial resolution images. Data fusion of multi-resolution images within optical sensor data is vital for visual interpretation, but may not improve classification performance. In contrast, integration of optical and radar data did improve classification performance when the proper data fusion method was used. Of the classification algorithms available, the maximum likelihood classifier is still an important method for providing reasonably good accuracy, but nonparametric algorithms, such as classification tree analysis, has the potential to provide better results. However, they often require more time to achieve parametric optimization. Proper use of hierarchical-based methods is fundamental for developing accurate land use/cover classification, mainly from historical remotely sensed data.
Directory of Open Access Journals (Sweden)
Serafini Maria
2003-11-01
Full Text Available Abstract Background We describe the E-RFE method for gene ranking, which is useful for the identification of markers in the predictive classification of array data. The method supports a practical modeling scheme designed to avoid the construction of classification rules based on the selection of too small gene subsets (an effect known as the selection bias, in which the estimated predictive errors are too optimistic due to testing on samples already considered in the feature selection process. Results With E-RFE, we speed up the recursive feature elimination (RFE with SVM classifiers by eliminating chunks of uninteresting genes using an entropy measure of the SVM weights distribution. An optimal subset of genes is selected according to a two-strata model evaluation procedure: modeling is replicated by an external stratified-partition resampling scheme, and, within each run, an internal K-fold cross-validation is used for E-RFE ranking. Also, the optimal number of genes can be estimated according to the saturation of Zipf's law profiles. Conclusions Without a decrease of classification accuracy, E-RFE allows a speed-up factor of 100 with respect to standard RFE, while improving on alternative parametric RFE reduction strategies. Thus, a process for gene selection and error estimation is made practical, ensuring control of the selection bias, and providing additional diagnostic indicators of gene importance.
Directory of Open Access Journals (Sweden)
Lianqing Zhu
2018-01-01
Full Text Available In order to improve the classification accuracy of Chinese Salvia miltiorrhiza using near-infrared spectroscopy, a novel local variable selection strategy is thus proposed. Combining the strengths of the local algorithm and interval partial least squares, the spectra data have firstly been divided into several pairs of classes in sample direction and equidistant subintervals in variable direction. Then, a local classification model has been built, and the most proper spectral region has been selected based on the new evaluation criterion considering both classification error rate and best predictive ability under the leave-one-out cross validation scheme for each pair of classes. Finally, each observation can be assigned to belong to the class according to the statistical analysis of classification results of the local classification model built on selected variables. The performance of the proposed method was demonstrated through near-infrared spectra of cultivated or wild Salvia miltiorrhiza, which are collected from 8 geographical origins in 5 provinces of China. For comparison, soft independent modelling of class analogy and partial least squares discriminant analysis methods are, respectively, employed as the classification model. Experimental results showed that classification performance of the classification model with local variable selection was obvious better than that without variable selection.
Deep Galaxy: Classification of Galaxies based on Deep Convolutional Neural Networks
Khalifa, Nour Eldeen M.; Taha, Mohamed Hamed N.; Hassanien, Aboul Ella; Selim, I. M.
2017-01-01
In this paper, a deep convolutional neural network architecture for galaxies classification is presented. The galaxy can be classified based on its features into main three categories Elliptical, Spiral, and Irregular. The proposed deep galaxies architecture consists of 8 layers, one main convolutional layer for features extraction with 96 filters, followed by two principles fully connected layers for classification. It is trained over 1356 images and achieved 97.272% in testing accuracy. A c...
REAL-TIME INTELLIGENT MULTILAYER ATTACK CLASSIFICATION SYSTEM
Directory of Open Access Journals (Sweden)
T. Subbhulakshmi
2014-01-01
Full Text Available Intrusion Detection Systems (IDS takes the lion’s share of the current security infrastructure. Detection of intrusions is vital for initiating the defensive procedures. Intrusion detection was done by statistical and distance based methods. A threshold value is used in these methods to indicate the level of normalcy. When the network traffic crosses the level of normalcy then above which it is flagged as anomalous. When there are occurrences of new intrusion events which are increasingly a key part of system security, the statistical techniques cannot detect them. To overcome this issue, learning techniques are used which helps in identifying new intrusion activities in a computer system. The objective of the proposed system designed in this paper is to classify the intrusions using an Intelligent Multi Layered Attack Classification System (IMLACS which helps in detecting and classifying the intrusions with improved classification accuracy. The intelligent multi layered approach contains three intelligent layers. The first layer involves Binary Support Vector Machine classification for detecting the normal and attack. The second layer involves neural network classification to classify the attacks into classes of attacks. The third layer involves fuzzy inference system to classify the attacks into various subclasses. The proposed IMLACS can be able to detect an intrusion behavior of the networks since the system contains a three intelligent layer classification and better set of rules. Feature selection is also used to improve the time of detection. The experimental results show that the IMLACS achieves the Classification Rate of 97.31%.
Advanced Steel Microstructural Classification by Deep Learning Methods.
Azimi, Seyed Majid; Britz, Dominik; Engstler, Michael; Fritz, Mario; Mücklich, Frank
2018-02-01
The inner structure of a material is called microstructure. It stores the genesis of a material and determines all its physical and chemical properties. While microstructural characterization is widely spread and well known, the microstructural classification is mostly done manually by human experts, which gives rise to uncertainties due to subjectivity. Since the microstructure could be a combination of different phases or constituents with complex substructures its automatic classification is very challenging and only a few prior studies exist. Prior works focused on designed and engineered features by experts and classified microstructures separately from the feature extraction step. Recently, Deep Learning methods have shown strong performance in vision applications by learning the features from data together with the classification step. In this work, we propose a Deep Learning method for microstructural classification in the examples of certain microstructural constituents of low carbon steel. This novel method employs pixel-wise segmentation via Fully Convolutional Neural Network (FCNN) accompanied by a max-voting scheme. Our system achieves 93.94% classification accuracy, drastically outperforming the state-of-the-art method of 48.89% accuracy. Beyond the strong performance of our method, this line of research offers a more robust and first of all objective way for the difficult task of steel quality appreciation.
Directory of Open Access Journals (Sweden)
Casey Olives
Full Text Available Originally a binary classifier, Lot Quality Assurance Sampling (LQAS has proven to be a useful tool for classification of the prevalence of Schistosoma mansoni into multiple categories (≤10%, >10 and <50%, ≥50%, and semi-curtailed sampling has been shown to effectively reduce the number of observations needed to reach a decision. To date the statistical underpinnings for Multiple Category-LQAS (MC-LQAS have not received full treatment. We explore the analytical properties of MC-LQAS, and validate its use for the classification of S. mansoni prevalence in multiple settings in East Africa.We outline MC-LQAS design principles and formulae for operating characteristic curves. In addition, we derive the average sample number for MC-LQAS when utilizing semi-curtailed sampling and introduce curtailed sampling in this setting. We also assess the performance of MC-LQAS designs with maximum sample sizes of n=15 and n=25 via a weighted kappa-statistic using S. mansoni data collected in 388 schools from four studies in East Africa.Overall performance of MC-LQAS classification was high (kappa-statistic of 0.87. In three of the studies, the kappa-statistic for a design with n=15 was greater than 0.75. In the fourth study, where these designs performed poorly (kappa-statistic less than 0.50, the majority of observations fell in regions where potential error is known to be high. Employment of semi-curtailed and curtailed sampling further reduced the sample size by as many as 0.5 and 3.5 observations per school, respectively, without increasing classification error.This work provides the needed analytics to understand the properties of MC-LQAS for assessing the prevalance of S. mansoni and shows that in most settings a sample size of 15 children provides a reliable classification of schools.
Search and Classification Using Multiple Autonomous Vehicles Decision-Making and Sensor Management
Wang, Yue
2012-01-01
Search and Classification Using Multiple Autonomous Vehicles provides a comprehensive study of decision-making strategies for domain search and object classification using multiple autonomous vehicles (MAV) under both deterministic and probabilistic frameworks. It serves as a first discussion of the problem of effective resource allocation using MAV with sensing limitations, i.e., for search and classification missions over large-scale domains, or when there are far more objects to be found and classified than there are autonomous vehicles available. Under such scenarios, search and classification compete for limited sensing resources. This is because search requires vehicle mobility while classification restricts the vehicles to the vicinity of any objects found. The authors develop decision-making strategies to choose between these competing tasks and vehicle-motion-control laws to achieve the proposed management scheme. Deterministic Lyapunov-based, probabilistic Bayesian-based, and risk-based decision-mak...
Classification of Clouds in Satellite Imagery Using Adaptive Fuzzy Sparse Representation
Directory of Open Access Journals (Sweden)
Wei Jin
2016-12-01
Full Text Available Automatic cloud detection and classification using satellite cloud imagery have various meteorological applications such as weather forecasting and climate monitoring. Cloud pattern analysis is one of the research hotspots recently. Since satellites sense the clouds remotely from space, and different cloud types often overlap and convert into each other, there must be some fuzziness and uncertainty in satellite cloud imagery. Satellite observation is susceptible to noises, while traditional cloud classification methods are sensitive to noises and outliers; it is hard for traditional cloud classification methods to achieve reliable results. To deal with these problems, a satellite cloud classification method using adaptive fuzzy sparse representation-based classification (AFSRC is proposed. Firstly, by defining adaptive parameters related to attenuation rate and critical membership, an improved fuzzy membership is introduced to accommodate the fuzziness and uncertainty of satellite cloud imagery; secondly, by effective combination of the improved fuzzy membership function and sparse representation-based classification (SRC, atoms in training dictionary are optimized; finally, an adaptive fuzzy sparse representation classifier for cloud classification is proposed. Experiment results on FY-2G satellite cloud image show that, the proposed method not only improves the accuracy of cloud classification, but also has strong stability and adaptability with high computational efficiency.
Classification of Clouds in Satellite Imagery Using Adaptive Fuzzy Sparse Representation
Jin, Wei; Gong, Fei; Zeng, Xingbin; Fu, Randi
2016-01-01
Automatic cloud detection and classification using satellite cloud imagery have various meteorological applications such as weather forecasting and climate monitoring. Cloud pattern analysis is one of the research hotspots recently. Since satellites sense the clouds remotely from space, and different cloud types often overlap and convert into each other, there must be some fuzziness and uncertainty in satellite cloud imagery. Satellite observation is susceptible to noises, while traditional cloud classification methods are sensitive to noises and outliers; it is hard for traditional cloud classification methods to achieve reliable results. To deal with these problems, a satellite cloud classification method using adaptive fuzzy sparse representation-based classification (AFSRC) is proposed. Firstly, by defining adaptive parameters related to attenuation rate and critical membership, an improved fuzzy membership is introduced to accommodate the fuzziness and uncertainty of satellite cloud imagery; secondly, by effective combination of the improved fuzzy membership function and sparse representation-based classification (SRC), atoms in training dictionary are optimized; finally, an adaptive fuzzy sparse representation classifier for cloud classification is proposed. Experiment results on FY-2G satellite cloud image show that, the proposed method not only improves the accuracy of cloud classification, but also has strong stability and adaptability with high computational efficiency. PMID:27999261
Zheng, Ling; Yumak, Hasan; Chen, Ling; Ochs, Christopher; Geller, James; Kapusnik-Uner, Joan; Perl, Yehoshua
2017-09-01
The National Drug File - Reference Terminology (NDF-RT) is a large and complex drug terminology consisting of several classification hierarchies on top of an extensive collection of drug concepts. These hierarchies provide important information about clinical drugs, e.g., their chemical ingredients, mechanisms of action, dosage form and physiological effects. Within NDF-RT such information is represented using tens of thousands of roles connecting drugs to classifications. In previous studies, we have introduced various kinds of Abstraction Networks to summarize the content and structure of terminologies in order to facilitate their visual comprehension, and support quality assurance of terminologies. However, these previous kinds of Abstraction Networks are not appropriate for summarizing the NDF-RT classification hierarchies, due to its unique structure. In this paper, we present the novel Ingredient Abstraction Network (IAbN) to summarize, visualize and support the audit of NDF-RT's Chemical Ingredients hierarchy and its associated drugs. A common theme in our quality assurance framework is to use characterizations of sets of concepts, revealed by the Abstraction Network structure, to capture concepts, the modeling of which is more complex than for other concepts. For the IAbN, we characterize drug ingredient concepts as more complex if they belong to IAbN groups with multiple parent groups. We show that such concepts have a statistically significantly higher rate of errors than a control sample and identify two especially common patterns of errors. Copyright © 2017 Elsevier Inc. All rights reserved.
A New Classification Technique in Mobile Robot Navigation
Directory of Open Access Journals (Sweden)
Bambang Tutuko
2011-12-01
Full Text Available This paper presents a novel pattern recognition algorithm that use weightless neural network (WNNs technique.This technique plays a role of situation classifier to judge the situation around the mobile robot environment and makes control decision in mobile robot navigation. The WNNs technique is choosen due to significant advantages over conventional neural network, such as they can be easily implemented in hardware using standard RAM, faster in training phase and work with small resources. Using a simple classification algorithm, the similar data will be grouped with each other and it will be possible to attach similar data classes to specific local areas in the mobile robot environment. This strategy is demonstrated in simple mobile robot powered by low cost microcontrollers with 512 bytes of RAM and low cost sensors. Experimental result shows, when number of neuron increases the average environmental recognition ratehas risen from 87.6% to 98.5%.The WNNs technique allows the mobile robot to recognize many and different environmental patterns and avoid obstacles in real time. Moreover, by using proposed WNNstechnique mobile robot has successfully reached the goal in dynamic environment compare to fuzzy logic technique and logic function, capable of dealing with uncertainty in sensor reading, achieving good performance in performing control actions with 0.56% error rate in mobile robot speed.
FEM for time-fractional diffusion equations, novel optimal error analyses
Mustapha, Kassem
2016-01-01
A semidiscrete Galerkin finite element method applied to time-fractional diffusion equations with time-space dependent diffusivity on bounded convex spatial domains will be studied. The main focus is on achieving optimal error results with respect to both the convergence order of the approximate solution and the regularity of the initial data. By using novel energy arguments, for each fixed time $t$, optimal error bounds in the spatial $L^2$- and $H^1$-norms are derived for both cases: smooth...
Simultaneous feature selection and classification via Minimax Probability Machine
Directory of Open Access Journals (Sweden)
Liming Yang
2010-12-01
Full Text Available This paper presents a novel method for simultaneous feature selection and classification by incorporating a robust L1-norm into the objective function of Minimax Probability Machine (MPM. A fractional programming framework is derived by using a bound on the misclassification error involving the mean and covariance of the data. Furthermore, the problems are solved by the Quadratic Interpolation method. Experiments show that our methods can select fewer features to improve the generalization compared to MPM, which illustrates the effectiveness of the proposed algorithms.
DEFF Research Database (Denmark)
Kheir, Rania Bou; Greve, Mogens Humlekrog; Bøcher, Peder Klith
2010-01-01
the geographic distribution of SOC across Denmark using remote sensing (RS), geographic information systems (GISs) and decision-tree modeling (un-pruned and pruned classification trees). Seventeen parameters, i.e. parent material, soil type, landscape type, elevation, slope gradient, slope aspect, mean curvature...... field measurements in the area of interest (Denmark). A large number of tree-based classification models (588) were developed using (i) all of the parameters, (ii) all Digital Elevation Model (DEM) parameters only, (iii) the primary DEM parameters only, (iv), the remote sensing (RS) indices only, (v......) selected pairs of parameters, (vi) soil type, parent material and landscape type only, and (vii) the parameters having a high impact on SOC distribution in built pruned trees. The best constructed classification tree models (in the number of three) with the lowest misclassification error (ME...
Dynamic classification system in large-scale supervision of energy efficiency in buildings
International Nuclear Information System (INIS)
Kiluk, S.
2014-01-01
Highlights: • Rough set approximation of classification improves energy efficiency prediction. • Dynamic features of diagnostic classification allow for its precise prediction. • Indiscernibility in large population enhances identification of process features. • Diagnostic information can be refined by dynamic references to local neighbourhood. • We introduce data exploration validation based on system dynamics and uncertainty. - Abstract: Data mining and knowledge discovery applied to the billing data provide the diagnostic instruments for the evaluation of energy use in buildings connected to a district heating network. To ensure the validity of an algorithm-based classification system, the dynamic properties of a sequence of partitions for consecutive detected events were investigated. The information regarding the dynamic properties of the classification system refers to the similarities between the supervised objects and migrations that originate from the changes in the building energy use and loss similarity to their neighbourhood and thus represents the refinement of knowledge. In this study, we demonstrate that algorithm-based diagnostic knowledge has dynamic properties that can be exploited with a rough set predictor to evaluate whether the implementation of classification for supervision of energy use aligns with the dynamics of changes of district heating-supplied building properties. Moreover, we demonstrate the refinement of the current knowledge with the previous findings and we present the creation of predictive diagnostic systems based on knowledge dynamics with a satisfactory level of classification errors, even for non-stationary data
Wang, Bingjie; Pi, Shaohua; Sun, Qi; Jia, Bo
2015-05-01
An improved classification algorithm that considers multiscale wavelet packet Shannon entropy is proposed. Decomposition coefficients at all levels are obtained to build the initial Shannon entropy feature vector. After subtracting the Shannon entropy map of the background signal, components of the strongest discriminating power in the initial feature vector are picked out to rebuild the Shannon entropy feature vector, which is transferred to radial basis function (RBF) neural network for classification. Four types of man-made vibrational intrusion signals are recorded based on a modified Sagnac interferometer. The performance of the improved classification algorithm has been evaluated by the classification experiments via RBF neural network under different diffusion coefficients. An 85% classification accuracy rate is achieved, which is higher than the other common algorithms. The classification results show that this improved classification algorithm can be used to classify vibrational intrusion signals in an automatic real-time monitoring system.
78 FR 68983 - Cotton Futures Classification: Optional Classification Procedure
2013-11-18
...-AD33 Cotton Futures Classification: Optional Classification Procedure AGENCY: Agricultural Marketing... regulations to allow for the addition of an optional cotton futures classification procedure--identified and... response to requests from the U.S. cotton industry and ICE, AMS will offer a futures classification option...
Alom, Md. Zahangir; Awwal, Abdul A. S.; Lowe-Webb, Roger; Taha, Tarek M.
2017-08-01
Deep-learning methods are gaining popularity because of their state-of-the-art performance in image classification tasks. In this paper, we explore classification of laser-beam images from the National Ignition Facility (NIF) using a novel deeplearning approach. NIF is the world's largest, most energetic laser. It has nearly 40,000 optics that precisely guide, reflect, amplify, and focus 192 laser beams onto a fusion target. NIF utilizes four petawatt lasers called the Advanced Radiographic Capability (ARC) to produce backlighting X-ray illumination to capture implosion dynamics of NIF experiments with picosecond temporal resolution. In the current operational configuration, four independent short-pulse ARC beams are created and combined in a split-beam configuration in each of two NIF apertures at the entry of the pre-amplifier. The subaperture beams then propagate through the NIF beampath up to the ARC compressor. Each ARC beamlet is separately compressed with a dedicated set of four gratings and recombined as sub-apertures for transport to the parabola vessel, where the beams are focused using parabolic mirrors and pointed to the target. Small angular errors in the compressor gratings can cause the sub-aperture beams to diverge from one another and prevent accurate alignment through the transport section between the compressor and parabolic mirrors. This is an off-normal condition that must be detected and corrected. The goal of the off-normal check is to determine whether the ARC beamlets are sufficiently overlapped into a merged single spot or diverged into two distinct spots. Thus, the objective of the current work is three-fold: developing a simple algorithm to perform off-normal classification, exploring the use of Convolutional Neural Network (CNN) for the same task, and understanding the inter-relationship of the two approaches. The CNN recognition results are compared with other machine-learning approaches, such as Deep Neural Network (DNN) and Support
Classification of deadlift biomechanics with wearable inertial measurement units.
O'Reilly, Martin A; Whelan, Darragh F; Ward, Tomas E; Delahunt, Eamonn; Caulfield, Brian M
2017-06-14
The deadlift is a compound full-body exercise that is fundamental in resistance training, rehabilitation programs and powerlifting competitions. Accurate quantification of deadlift biomechanics is important to reduce the risk of injury and ensure training and rehabilitation goals are achieved. This study sought to develop and evaluate deadlift exercise technique classification systems utilising Inertial Measurement Units (IMUs), recording at 51.2Hz, worn on the lumbar spine, both thighs and both shanks. It also sought to compare classification quality when these IMUs are worn in combination and in isolation. Two datasets of IMU deadlift data were collected. Eighty participants first completed deadlifts with acceptable technique and 5 distinct, deliberately induced deviations from acceptable form. Fifty-five members of this group also completed a fatiguing protocol (3-Repition Maximum test) to enable the collection of natural deadlift deviations. For both datasets, universal and personalised random-forests classifiers were developed and evaluated. Personalised classifiers outperformed universal classifiers in accuracy, sensitivity and specificity in the binary classification of acceptable or aberrant technique and in the multi-label classification of specific deadlift deviations. Whilst recent research has favoured universal classifiers due to the reduced overhead in setting them up for new system users, this work demonstrates that such techniques may not be appropriate for classifying deadlift technique due to the poor accuracy achieved. However, personalised classifiers perform very well in assessing deadlift technique, even when using data derived from a single lumbar-worn IMU to detect specific naturally occurring technique mistakes. Copyright © 2017 Elsevier Ltd. All rights reserved.
Concussion classification via deep learning using whole-brain white matter fiber strains
Cai, Yunliang; Wu, Shaoju; Zhao, Wei; Li, Zhigang; Wu, Zheyang
2018-01-01
Developing an accurate and reliable injury predictor is central to the biomechanical studies of traumatic brain injury. State-of-the-art efforts continue to rely on empirical, scalar metrics based on kinematics or model-estimated tissue responses explicitly pre-defined in a specific brain region of interest. They could suffer from loss of information. A single training dataset has also been used to evaluate performance but without cross-validation. In this study, we developed a deep learning approach for concussion classification using implicit features of the entire voxel-wise white matter fiber strains. Using reconstructed American National Football League (NFL) injury cases, leave-one-out cross-validation was employed to objectively compare injury prediction performances against two baseline machine learning classifiers (support vector machine (SVM) and random forest (RF)) and four scalar metrics via univariate logistic regression (Brain Injury Criterion (BrIC), cumulative strain damage measure of the whole brain (CSDM-WB) and the corpus callosum (CSDM-CC), and peak fiber strain in the CC). Feature-based machine learning classifiers including deep learning, SVM, and RF consistently outperformed all scalar injury metrics across all performance categories (e.g., leave-one-out accuracy of 0.828–0.862 vs. 0.690–0.776, and .632+ error of 0.148–0.176 vs. 0.207–0.292). Further, deep learning achieved the best cross-validation accuracy, sensitivity, AUC, and .632+ error. These findings demonstrate the superior performances of deep learning in concussion prediction and suggest its promise for future applications in biomechanical investigations of traumatic brain injury. PMID:29795640
Concussion classification via deep learning using whole-brain white matter fiber strains.
Cai, Yunliang; Wu, Shaoju; Zhao, Wei; Li, Zhigang; Wu, Zheyang; Ji, Songbai
2018-01-01
Developing an accurate and reliable injury predictor is central to the biomechanical studies of traumatic brain injury. State-of-the-art efforts continue to rely on empirical, scalar metrics based on kinematics or model-estimated tissue responses explicitly pre-defined in a specific brain region of interest. They could suffer from loss of information. A single training dataset has also been used to evaluate performance but without cross-validation. In this study, we developed a deep learning approach for concussion classification using implicit features of the entire voxel-wise white matter fiber strains. Using reconstructed American National Football League (NFL) injury cases, leave-one-out cross-validation was employed to objectively compare injury prediction performances against two baseline machine learning classifiers (support vector machine (SVM) and random forest (RF)) and four scalar metrics via univariate logistic regression (Brain Injury Criterion (BrIC), cumulative strain damage measure of the whole brain (CSDM-WB) and the corpus callosum (CSDM-CC), and peak fiber strain in the CC). Feature-based machine learning classifiers including deep learning, SVM, and RF consistently outperformed all scalar injury metrics across all performance categories (e.g., leave-one-out accuracy of 0.828-0.862 vs. 0.690-0.776, and .632+ error of 0.148-0.176 vs. 0.207-0.292). Further, deep learning achieved the best cross-validation accuracy, sensitivity, AUC, and .632+ error. These findings demonstrate the superior performances of deep learning in concussion prediction and suggest its promise for future applications in biomechanical investigations of traumatic brain injury.
Challenge and Error: Critical Events and Attention-Related Errors
Cheyne, James Allan; Carriere, Jonathan S. A.; Solman, Grayden J. F.; Smilek, Daniel
2011-01-01
Attention lapses resulting from reactivity to task challenges and their consequences constitute a pervasive factor affecting everyday performance errors and accidents. A bidirectional model of attention lapses (error [image omitted] attention-lapse: Cheyne, Solman, Carriere, & Smilek, 2009) argues that errors beget errors by generating attention…
Error forecasting schemes of error correction at receiver
International Nuclear Information System (INIS)
Bhunia, C.T.
2007-08-01
To combat error in computer communication networks, ARQ (Automatic Repeat Request) techniques are used. Recently Chakraborty has proposed a simple technique called the packet combining scheme in which error is corrected at the receiver from the erroneous copies. Packet Combining (PC) scheme fails: (i) when bit error locations in erroneous copies are the same and (ii) when multiple bit errors occur. Both these have been addressed recently by two schemes known as Packet Reversed Packet Combining (PRPC) Scheme, and Modified Packet Combining (MPC) Scheme respectively. In the letter, two error forecasting correction schemes are reported, which in combination with PRPC offer higher throughput. (author)
The achievements and errors of a process of popular insurrection: Egypt, 2011-2014
Directory of Open Access Journals (Sweden)
José Carlos Castañeda Reyes
2018-04-01
Full Text Available The purpose of this paper is to evaluate, from a historical perspective, the major achievements of a process of popular insurrection in Egypt beginning on January 25, 2011, which concluded with the promulgation of the new Egyptian Constitution in January, 2014, while evaluating as well its notable mistakes. These were four years of intense historical reality, when the Egyptian people were able to carry out what was called the thawratâni masriyyatâni, two “Egyptian revolutions” that tried to achieve the objectives of “bread, freedom, social justice and human dignity”—which became the goals for the progress of rebellion witnessed by the world during that period. This was a time of significant achievements but also very costly mistakes that allowed the coming to power of a regime where there was military control over the popular forces that made the 2011 and 2013 “revolutions”—military control with the aim of preventing the implementation of measures for real—economic, social, political—change in Egypt.
Energy Technology Data Exchange (ETDEWEB)
Kim, Sa Kil; Jang, Tong Il; Lee, Yong Hee; Shin, Kwang Hyeon [KAERI, Daejeon (Korea, Republic of)
2012-10-15
According to the Operational Performance Information System (OPIS) which has been operated to improve the public understanding by the KINS (Korea Institute of Nuclear Safety), unintended trip events by mainly human errors counted up to 38 cases (18.7%) from 2000 to 2011. Although the Nuclear Power Plant (NPP) industry in Korea has been making efforts to reduce the human errors which have largely contributed to trip events, the human error rate might keep increasing. Interestingly, digital based I and C systems is the one of the reduction factors of unintended reactor trips. Human errors, however, have occurred due to the digital based I and C systems because those systems require new or changed behaviors to the NPP operators. Therefore, it is necessary that the investigations of human errors consider a new methodology to find not only tangible behavior but also intangible behavior such as organizational behaviors. In this study we investigated human errors to find latent factors such as decisions and conditions in the all of the unintended reactor trip events during last dozen years. To find them, we applied the HFACS (Human Factors Analysis and Classification System) which is a commonly utilized tool for investigating human contributions to aviation accidents under a widespread evaluation scheme. The objective of this study is to find latent factors behind of human errors in nuclear reactor trip events. Therefore, a method to investigate unintended trip events by human errors and the results will be discussed in more detail.
International Nuclear Information System (INIS)
Kim, Sa Kil; Jang, Tong Il; Lee, Yong Hee; Shin, Kwang Hyeon
2012-01-01
According to the Operational Performance Information System (OPIS) which has been operated to improve the public understanding by the KINS (Korea Institute of Nuclear Safety), unintended trip events by mainly human errors counted up to 38 cases (18.7%) from 2000 to 2011. Although the Nuclear Power Plant (NPP) industry in Korea has been making efforts to reduce the human errors which have largely contributed to trip events, the human error rate might keep increasing. Interestingly, digital based I and C systems is the one of the reduction factors of unintended reactor trips. Human errors, however, have occurred due to the digital based I and C systems because those systems require new or changed behaviors to the NPP operators. Therefore, it is necessary that the investigations of human errors consider a new methodology to find not only tangible behavior but also intangible behavior such as organizational behaviors. In this study we investigated human errors to find latent factors such as decisions and conditions in the all of the unintended reactor trip events during last dozen years. To find them, we applied the HFACS (Human Factors Analysis and Classification System) which is a commonly utilized tool for investigating human contributions to aviation accidents under a widespread evaluation scheme. The objective of this study is to find latent factors behind of human errors in nuclear reactor trip events. Therefore, a method to investigate unintended trip events by human errors and the results will be discussed in more detail
Applying Active Learning to Assertion Classification of Concepts in Clinical Text
Chen, Yukun; Mani, Subramani; Xu, Hua
2012-01-01
Supervised machine learning methods for clinical natural language processing (NLP) research require a large number of annotated samples, which are very expensive to build because of the involvement of physicians. Active learning, an approach that actively samples from a large pool, provides an alternative solution. Its major goal in classification is to reduce the annotation effort while maintaining the quality of the predictive model. However, few studies have investigated its uses in clinical NLP. This paper reports an application of active learning to a clinical text classification task: to determine the assertion status of clinical concepts. The annotated corpus for the assertion classification task in the 2010 i2b2/VA Clinical NLP Challenge was used in this study. We implemented several existing and newly developed active learning algorithms and assessed their uses. The outcome is reported in the global ALC score, based on the Area under the average Learning Curve of the AUC (Area Under the Curve) score. Results showed that when the same number of annotated samples was used, active learning strategies could generate better classification models (best ALC – 0.7715) than the passive learning method (random sampling) (ALC – 0.7411). Moreover, to achieve the same classification performance, active learning strategies required fewer samples than the random sampling method. For example, to achieve an AUC of 0.79, the random sampling method used 32 samples, while our best active learning algorithm required only 12 samples, a reduction of 62.5% in manual annotation effort. PMID:22127105
A novel application of deep learning for single-lead ECG classification.
Mathews, Sherin M; Kambhamettu, Chandra; Barner, Kenneth E
2018-06-04
Detecting and classifying cardiac arrhythmias is critical to the diagnosis of patients with cardiac abnormalities. In this paper, a novel approach based on deep learning methodology is proposed for the classification of single-lead electrocardiogram (ECG) signals. We demonstrate the application of the Restricted Boltzmann Machine (RBM) and deep belief networks (DBN) for ECG classification following detection of ventricular and supraventricular heartbeats using single-lead ECG. The effectiveness of this proposed algorithm is illustrated using real ECG signals from the widely-used MIT-BIH database. Simulation results demonstrate that with a suitable choice of parameters, RBM and DBN can achieve high average recognition accuracies of ventricular ectopic beats (93.63%) and of supraventricular ectopic beats (95.57%) at a low sampling rate of 114 Hz. Experimental results indicate that classifiers built into this deep learning-based framework achieved state-of-the art performance models at lower sampling rates and simple features when compared to traditional methods. Further, employing features extracted at a sampling rate of 114 Hz when combined with deep learning provided enough discriminatory power for the classification task. This performance is comparable to that of traditional methods and uses a much lower sampling rate and simpler features. Thus, our proposed deep neural network algorithm demonstrates that deep learning-based methods offer accurate ECG classification and could potentially be extended to other physiological signal classifications, such as those in arterial blood pressure (ABP), nerve conduction (EMG), and heart rate variability (HRV) studies. Copyright © 2018. Published by Elsevier Ltd.
Directory of Open Access Journals (Sweden)
Joseph St. Peter
2018-03-01
Full Text Available Land cover classification provides valuable information for prioritizing management and conservation operations across large landscapes. Current regional scale land cover geospatial products within the United States have a spatial resolution that is too coarse to provide the necessary information for operations at the local and project scales. This paper describes a methodology that uses recent advances in spatial analysis software to create a land cover classification over a large region in the southeastern United States at a fine (1 m spatial resolution. This methodology used image texture metrics and principle components derived from National Agriculture Imagery Program (NAIP aerial photographic imagery, visually classified locations, and a softmax neural network model. The model efficiently produced classification surfaces at 1 m resolution across roughly 11.6 million hectares (28.8 million acres with less than 10% average error in modeled probability. The classification surfaces consist of probability estimates of 13 visually distinct classes for each 1 m cell across the study area. This methodology and the tools used in this study constitute a highly flexible fine resolution land cover classification that can be applied across large extents using standard computer hardware, common and open source software and publicly available imagery.
Complications in Endovascular Neurosurgery: Critical Analysis and Classification.
Ravindra, Vijay M; Mazur, Marcus D; Park, Min S; Kilburg, Craig; Moran, Christopher J; Hardman, Rulon L; Couldwell, William T; Taussky, Philipp
2016-11-01
Precisely defining complications, which are used to measure overall quality, is necessary for critical review of delivery of care and quality improvement in endovascular neurosurgery, which lacks common definitions for complications. Furthermore, in endovascular interventions, events that may be labeled complications may not always negatively affect outcome. Our objective is to provide precise definitions for quality evaluation within endovascular neurosurgery. Thus, we propose an endovascular-specific classification system of complications based on our own patient series. This single-center review included all patients who had endovascular interventions from September 2013 to August 2015. Complication types were analyzed, and a descriptive analysis was undertaken to calculate the incidence of complications overall and in each category. Two hundred and seventy-five endovascular interventions were performed in 245 patients (65% female; mean age, 55 years). Forty complications occurred in 39 patients (15%), most commonly during treatment of intracranial aneurysms (24/40). Mechanical complications (eg, device deployment, catheter, or closure device failure) occurred in 8/40, technical complications (eg, failure to deploy flow diverter, unintended embolization, air emboli, retroperitoneal hemorrhage, dissection) in 11/40, judgment errors (eg, patient or equipment selection) in 9/40, and critical events (eg, groin hematoma, hemorrhagic or thromboembolic complications) in 12/40 patients. Only 12/40 complications (30%) resulted in new neurologic deficits, vessel injury requiring surgery, or blood transfusion. We propose an endovascular-specific classification system of complications with 4 categories: mechanical, technical, judgment errors, and critical events. This system provides a framework for future studies and quality control in endovascular neurosurgery. Copyright © 2016 Elsevier Inc. All rights reserved.
A proposal of criteria for the classification of systemic sclerosis.
Nadashkevich, Oleg; Davis, Paul; Fritzler, Marvin J
2004-11-01
Sensitive and specific criteria for the classification of systemic sclerosis are required by clinicians and investigators to achieve higher quality clinical studies and approaches to therapy. A clinical study of systemic sclerosis patients in Europe and Canada led to a set of criteria that achieve high sensitivity and specificity. Both clinical and laboratory investigations of patients with systemic sclerosis, related conditions and diseases with clinical features that can be mistaken as part of the systemic sclerosis spectrum were undertaken. Laboratory investigations included the detection of autoantibodies to centromere proteins, Scl-70 (topoisomerase I), and fibrillarin (U3-RNP). Based on the investigation of 269 systemic sclerosis patients and 720 patients presenting with related and confounding conditions, the following set of criteria for the classification of systemic sclerosis was proposed: 1) autoantibodies to: centromere proteins, Scl-70 (topo I), fibrillarin; 2) bibasilar pulmonary fibrosis; 3) contractures of the digital joints or prayer sign; 4) dermal thickening proximal to the wrists; 5) calcinosis cutis; 6) Raynaud's phenomenon; 7) esophageal distal hypomotility or reflux-esophagitis; 8) sclerodactyly or non-pitting digital edema; 9) teleangiectasias. The classification of definite SSc requires at least three of the above criteria. Criteria for the classification of systemic sclerosis have been proposed. Preliminary testing has defined the sensitivity and specificity of these criteria as high as 99% and 100%, respectively. Testing and validation of the proposed criteria by other clinical centers is required.
Classification Accuracy Increase Using Multisensor Data Fusion
Makarau, A.; Palubinskas, G.; Reinartz, P.
2011-09-01
The practical use of very high resolution visible and near-infrared (VNIR) data is still growing (IKONOS, Quickbird, GeoEye-1, etc.) but for classification purposes the number of bands is limited in comparison to full spectral imaging. These limitations may lead to the confusion of materials such as different roofs, pavements, roads, etc. and therefore may provide wrong interpretation and use of classification products. Employment of hyperspectral data is another solution, but their low spatial resolution (comparing to multispectral data) restrict their usage for many applications. Another improvement can be achieved by fusion approaches of multisensory data since this may increase the quality of scene classification. Integration of Synthetic Aperture Radar (SAR) and optical data is widely performed for automatic classification, interpretation, and change detection. In this paper we present an approach for very high resolution SAR and multispectral data fusion for automatic classification in urban areas. Single polarization TerraSAR-X (SpotLight mode) and multispectral data are integrated using the INFOFUSE framework, consisting of feature extraction (information fission), unsupervised clustering (data representation on a finite domain and dimensionality reduction), and data aggregation (Bayesian or neural network). This framework allows a relevant way of multisource data combination following consensus theory. The classification is not influenced by the limitations of dimensionality, and the calculation complexity primarily depends on the step of dimensionality reduction. Fusion of single polarization TerraSAR-X, WorldView-2 (VNIR or full set), and Digital Surface Model (DSM) data allow for different types of urban objects to be classified into predefined classes of interest with increased accuracy. The comparison to classification results of WorldView-2 multispectral data (8 spectral bands) is provided and the numerical evaluation of the method in comparison to
International Nuclear Information System (INIS)
Knuefer; Lindauer
1980-01-01
Besides that at spectacular events a combination of component failure and human error is often found. Especially the Rasmussen-Report and the German Risk Assessment Study show for pressurised water reactors that human error must not be underestimated. Although operator errors as a form of human error can never be eliminated entirely, they can be minimized and their effects kept within acceptable limits if a thorough training of personnel is combined with an adequate design of the plant against accidents. Contrary to the investigation of engineering errors, the investigation of human errors has so far been carried out with relatively small budgets. Intensified investigations in this field appear to be a worthwhile effort. (orig.)
Comparing Interval Management Control Laws for Steady-State Errors and String Stability
Weitz, Lesley A.; Swieringa, Kurt A.
2018-01-01
Interval Management (IM) is a future airborne spacing concept that leverages avionics to provide speed guidance to an aircraft to achieve and maintain a specified spacing interval from another aircraft. The design of a speed control law to achieve the spacing goal is a key aspect in the research and development of the IM concept. In this paper, two control laws that are used in much of the contemporary IM research are analyzed and compared to characterize steady-state errors and string stability. Numerical results are used to illustrate how the choice of control laws gains impacts the size of steady-state errors and string performance and the potential trade-offs between those performance characteristics.
Paletz, Susannah B F; Bearman, Christopher; Orasanu, Judith; Holbrook, Jon
2009-08-01
The presence of social psychological pressures on pilot decision making was assessed using qualitative analyses of critical incident interviews. Social psychological phenomena have long been known to influence attitudes and behavior but have not been highlighted in accident investigation models. Using a critical incident method, 28 pilots who flew in Alaska were interviewed. The participants were asked to describe a situation involving weather when they were pilot in command and found their skills challenged. They were asked to describe the incident in detail but were not explicitly asked to identify social pressures. Pressures were extracted from transcripts in a bottom-up manner and then clustered into themes. Of the 28 pilots, 16 described social psychological pressures on their decision making, specifically, informational social influence, the foot-in-the-door persuasion technique, normalization of deviance, and impression management and self-consistency motives. We believe accident and incident investigations can benefit from explicit inclusion of common social psychological pressures. We recommend specific ways of incorporating these pressures into theHuman Factors Analysis and Classification System.
Ship Detection and Classification on Optical Remote Sensing Images Using Deep Learning
Directory of Open Access Journals (Sweden)
Liu Ying
2017-01-01
Full Text Available Ship detection and classification is critical for national maritime security and national defense. Although some SAR (Synthetic Aperture Radar image-based ship detection approaches have been proposed and used, they are not able to satisfy the requirement of real-world applications as the number of SAR sensors is limited, the resolution is low, and the revisit cycle is long. As massive optical remote sensing images of high resolution are available, ship detection and classification on theses images is becoming a promising technique, and has attracted great attention on applications including maritime security and traffic control. Some digital image processing methods have been proposed to detect ships in optical remote sensing images, but most of them face difficulty in terms of accuracy, performance and complexity. Recently, an autoencoder-based deep neural network with extreme learning machine was proposed, but it cannot meet the requirement of real-world applications as it only works with simple and small-scaled data sets. Therefore, in this paper, we propose a novel ship detection and classification approach which utilizes deep convolutional neural network (CNN as the ship classifier. The performance of our proposed ship detection and classification approach was evaluated on a set of images downloaded from Google Earth at the resolution 0.5m. 99% detection accuracy and 95% classification accuracy were achieved. In model training, 75× speedup is achieved on 1 Nvidia Titanx GPU.
Random access to mobile networks with advanced error correction
Dippold, Michael
1990-01-01
A random access scheme for unreliable data channels is investigated in conjunction with an adaptive Hybrid-II Automatic Repeat Request (ARQ) scheme using Rate Compatible Punctured Codes (RCPC) Forward Error Correction (FEC). A simple scheme with fixed frame length and equal slot sizes is chosen and reservation is implicit by the first packet transmitted randomly in a free slot, similar to Reservation Aloha. This allows the further transmission of redundancy if the last decoding attempt failed. Results show that a high channel utilization and superior throughput can be achieved with this scheme that shows a quite low implementation complexity. For the example of an interleaved Rayleigh channel and soft decision utilization and mean delay are calculated. A utilization of 40 percent may be achieved for a frame with the number of slots being equal to half the station number under high traffic load. The effects of feedback channel errors and some countermeasures are discussed.
ERRORS AND DIFFICULTIES IN TRANSLATING LEGAL TEXTS
Directory of Open Access Journals (Sweden)
Camelia, CHIRILA
2014-11-01
Full Text Available Nowadays the accurate translation of legal texts has become highly important as the mistranslation of a passage in a contract, for example, could lead to lawsuits and loss of money. Consequently, the translation of legal texts to other languages faces many difficulties and only professional translators specialised in legal translation should deal with the translation of legal documents and scholarly writings. The purpose of this paper is to analyze translation from three perspectives: translation quality, errors and difficulties encountered in translating legal texts and consequences of such errors in professional translation. First of all, the paper points out the importance of performing a good and correct translation, which is one of the most important elements to be considered when discussing translation. Furthermore, the paper presents an overview of the errors and difficulties in translating texts and of the consequences of errors in professional translation, with applications to the field of law. The paper is also an approach to the differences between languages (English and Romanian that can hinder comprehension for those who have embarked upon the difficult task of translation. The research method that I have used to achieve the objectives of the paper was the content analysis of various Romanian and foreign authors' works.
Error Field Correction in DIII-D Ohmic Plasmas With Either Handedness
International Nuclear Information System (INIS)
Park, Jong-Kyu; Schaffer, Michael J.; La Haye, Robert J.; Scoville, Timothy J.; Menard, Jonathan E.
2011-01-01
Error field correction results in DIII-D plasmas are presented in various configurations. In both left-handed and right-handed plasma configurations, where the intrinsic error fields become different due to the opposite helical twist (handedness) of the magnetic field, the optimal error correction currents and the toroidal phases of internal(I)-coils are empirically established. Applications of the Ideal Perturbed Equilibrium Code to these results demonstrate that the field component to be minimized is not the resonant component of the external field, but the total field including ideal plasma responses. Consistency between experiment and theory has been greatly improved along with the understanding of ideal plasma responses, but non-ideal plasma responses still need to be understood to achieve the reliable predictability in tokamak error field correction.
Gender classification system in uncontrolled environments
Zeng, Pingping; Zhang, Yu-Jin; Duan, Fei
2011-01-01
Most face analysis systems available today perform mainly on restricted databases of images in terms of size, age, illumination. In addition, it is frequently assumed that all images are frontal and unconcealed. Actually, in a non-guided real-time supervision, the face pictures taken may often be partially covered and with head rotation less or more. In this paper, a special system supposed to be used in real-time surveillance with un-calibrated camera and non-guided photography is described. It mainly consists of five parts: face detection, non-face filtering, best-angle face selection, texture normalization, and gender classification. Emphases are focused on non-face filtering and best-angle face selection parts as well as texture normalization. Best-angle faces are figured out by PCA reconstruction, which equals to an implicit face alignment and results in a huge increase of the accuracy for gender classification. Dynamic skin model and a masked PCA reconstruction algorithm are applied to filter out faces detected in error. In order to fully include facial-texture and shape-outline features, a hybrid feature that is a combination of Gabor wavelet and PHoG (pyramid histogram of gradients) was proposed to equitable inner texture and outer contour. Comparative study on the effects of different non-face filtering and texture masking methods in the context of gender classification by SVM is reported through experiments on a set of UT (a company name) face images, a large number of internet images and CAS (Chinese Academy of Sciences) face database. Some encouraging results are obtained.
How Do Simulated Error Experiences Impact Attitudes Related to Error Prevention?
Breitkreuz, Karen R; Dougal, Renae L; Wright, Melanie C
2016-10-01
The objective of this project was to determine whether simulated exposure to error situations changes attitudes in a way that may have a positive impact on error prevention behaviors. Using a stratified quasi-randomized experiment design, we compared risk perception attitudes of a control group of nursing students who received standard error education (reviewed medication error content and watched movies about error experiences) to an experimental group of students who reviewed medication error content and participated in simulated error experiences. Dependent measures included perceived memorability of the educational experience, perceived frequency of errors, and perceived caution with respect to preventing errors. Experienced nursing students perceived the simulated error experiences to be more memorable than movies. Less experienced students perceived both simulated error experiences and movies to be highly memorable. After the intervention, compared with movie participants, simulation participants believed errors occurred more frequently. Both types of education increased the participants' intentions to be more cautious and reported caution remained higher than baseline for medication errors 6 months after the intervention. This study provides limited evidence of an advantage of simulation over watching movies describing actual errors with respect to manipulating attitudes related to error prevention. Both interventions resulted in long-term impacts on perceived caution in medication administration. Simulated error experiences made participants more aware of how easily errors can occur, and the movie education made participants more aware of the devastating consequences of errors.
Classification of Patients Treated for Infertility Using the IVF Method
Directory of Open Access Journals (Sweden)
Malinowski Paweł
2015-12-01
Full Text Available One of the most effective methods of infertility treatment is in vitro fertilization (IVF. Effectiveness of the treatment, as well as classification of the data obtained from it, is still an ongoing issue. Classifiers obtained so far are powerful, but even the best ones do not exhibit equal quality concerning possible treatment outcome predictions. Usually, lack of pregnancy is predicted far too often. This creates a constant need for further exploration of this issue. Careful use of different classification methods can, however, help to achieve that goal.
Underwater object classification using scattering transform of sonar signals
Saito, Naoki; Weber, David S.
2017-08-01
In this paper, we apply the scattering transform (ST)-a nonlinear map based off of a convolutional neural network (CNN)-to classification of underwater objects using sonar signals. The ST formalizes the observation that the filters learned by a CNN have wavelet-like structure. We achieve effective binary classification both on a real dataset of Unexploded Ordinance (UXOs), as well as synthetically generated examples. We also explore the effects on the waveforms with respect to changes in the object domain (e.g., translation, rotation, and acoustic impedance, etc.), and examine the consequences coming from theoretical results for the scattering transform. We show that the scattering transform is capable of excellent classification on both the synthetic and real problems, thanks to having more quasi-invariance properties that are well-suited to translation and rotation of the object.
Thermal error analysis and compensation for digital image/volume correlation
Pan, Bing
2018-02-01
Digital image/volume correlation (DIC/DVC) rely on the digital images acquired by digital cameras and x-ray CT scanners to extract the motion and deformation of test samples. Regrettably, these imaging devices are unstable optical systems, whose imaging geometry may undergo unavoidable slight and continual changes due to self-heating effect or ambient temperature variations. Changes in imaging geometry lead to both shift and expansion in the recorded 2D or 3D images, and finally manifest as systematic displacement and strain errors in DIC/DVC measurements. Since measurement accuracy is always the most important requirement in various experimental mechanics applications, these thermal-induced errors (referred to as thermal errors) should be given serious consideration in order to achieve high accuracy, reproducible DIC/DVC measurements. In this work, theoretical analyses are first given to understand the origin of thermal errors. Then real experiments are conducted to quantify thermal errors. Three solutions are suggested to mitigate or correct thermal errors. Among these solutions, a reference sample compensation approach is highly recommended because of its easy implementation, high accuracy and in-situ error correction capability. Most of the work has appeared in our previously published papers, thus its originality is not claimed. Instead, this paper aims to give a comprehensive overview and more insights of our work on thermal error analysis and compensation for DIC/DVC measurements.
Minimax bounds for active learning
Castro, R.M.; Nowak, R.; Bshouty, N.H.; Gentile, C.
2007-01-01
This paper aims to shed light on achievable limits in active learning. Using minimax analysis techniques, we study the achievable rates of classification error convergence for broad classes of distributions characterized by decision boundary regularity and noise conditions. The results clearly
Object Detection and Classification by Decision-Level Fusion for Intelligent Vehicle Systems
Directory of Open Access Journals (Sweden)
Sang-Il Oh
2017-01-01
Full Text Available To understand driving environments effectively, it is important to achieve accurate detection and classification of objects detected by sensor-based intelligent vehicle systems, which are significantly important tasks. Object detection is performed for the localization of objects, whereas object classification recognizes object classes from detected object regions. For accurate object detection and classification, fusing multiple sensor information into a key component of the representation and perception processes is necessary. In this paper, we propose a new object-detection and classification method using decision-level fusion. We fuse the classification outputs from independent unary classifiers, such as 3D point clouds and image data using a convolutional neural network (CNN. The unary classifiers for the two sensors are the CNN with five layers, which use more than two pre-trained convolutional layers to consider local to global features as data representation. To represent data using convolutional layers, we apply region of interest (ROI pooling to the outputs of each layer on the object candidate regions generated using object proposal generation to realize color flattening and semantic grouping for charge-coupled device and Light Detection And Ranging (LiDAR sensors. We evaluate our proposed method on a KITTI benchmark dataset to detect and classify three object classes: cars, pedestrians and cyclists. The evaluation results show that the proposed method achieves better performance than the previous methods. Our proposed method extracted approximately 500 proposals on a 1226 × 370 image, whereas the original selective search method extracted approximately 10 6 × n proposals. We obtained classification performance with 77.72% mean average precision over the entirety of the classes in the moderate detection level of the KITTI benchmark dataset.
Classification rates: non‐parametric verses parametric models using ...
African Journals Online (AJOL)
This research sought to establish if non parametric modeling achieves a higher correct classification ratio than a parametric model. The local likelihood technique was used to model fit the data sets. The same sets of data were modeled using parametric logit and the abilities of the two models to correctly predict the binary ...
Modified DCTNet for audio signals classification
Xian, Yin; Pu, Yunchen; Gan, Zhe; Lu, Liang; Thompson, Andrew
2016-10-01
In this paper, we investigate DCTNet for audio signal classification. Its output feature is related to Cohen's class of time-frequency distributions. We introduce the use of adaptive DCTNet (A-DCTNet) for audio signals feature extraction. The A-DCTNet applies the idea of constant-Q transform, with its center frequencies of filterbanks geometrically spaced. The A-DCTNet is adaptive to different acoustic scales, and it can better capture low frequency acoustic information that is sensitive to human audio perception than features such as Mel-frequency spectral coefficients (MFSC). We use features extracted by the A-DCTNet as input for classifiers. Experimental results show that the A-DCTNet and Recurrent Neural Networks (RNN) achieve state-of-the-art performance in bird song classification rate, and improve artist identification accuracy in music data. They demonstrate A-DCTNet's applicability to signal processing problems.
Toward functional classification of neuronal types.
Sharpee, Tatyana O
2014-09-17
How many types of neurons are there in the brain? This basic neuroscience question remains unsettled despite many decades of research. Classification schemes have been proposed based on anatomical, electrophysiological, or molecular properties. However, different schemes do not always agree with each other. This raises the question of whether one can classify neurons based on their function directly. For example, among sensory neurons, can a classification scheme be devised that is based on their role in encoding sensory stimuli? Here, theoretical arguments are outlined for how this can be achieved using information theory by looking at optimal numbers of cell types and paying attention to two key properties: correlations between inputs and noise in neural responses. This theoretical framework could help to map the hierarchical tree relating different neuronal classes within and across species. Copyright © 2014 Elsevier Inc. All rights reserved.
Round-off error in long-term orbital integrations using multistep methods
Quinlan, Gerald D.
1994-01-01
Techniques for reducing roundoff error are compared by testing them on high-order Stormer and summetric multistep methods. The best technique for most applications is to write the equation in summed, function-evaluation form and to store the coefficients as rational numbers. A larger error reduction can be achieved by writing the equation in backward-difference form and performing some of the additions in extended precision, but this entails a larger central processing unit (cpu) cost.
Error Immune Logic for Low-Power Probabilistic Computing
Directory of Open Access Journals (Sweden)
Bo Marr
2010-01-01
design for the maximum amount of energy savings per a given error rate. Spice simulation results using a commercially available and well-tested 0.25 μm technology are given verifying the ultra-low power, probabilistic full-adder designs. Further, close to 6X energy savings is achieved for a probabilistic full-adder over the deterministic case.
Is overall similarity classification less effortful than single-dimension classification?
Wills, Andy J; Milton, Fraser; Longmore, Christopher A; Hester, Sarah; Robinson, Jo
2013-01-01
It is sometimes argued that the implementation of an overall similarity classification is less effortful than the implementation of a single-dimension classification. In the current article, we argue that the evidence securely in support of this view is limited, and report additional evidence in support of the opposite proposition--overall similarity classification is more effortful than single-dimension classification. Using a match-to-standards procedure, Experiments 1A, 1B and 2 demonstrate that concurrent load reduces the prevalence of overall similarity classification, and that this effect is robust to changes in the concurrent load task employed, the level of time pressure experienced, and the short-term memory requirements of the classification task. Experiment 3 demonstrates that participants who produced overall similarity classifications from the outset have larger working memory capacities than those who produced single-dimension classifications initially, and Experiment 4 demonstrates that instructions to respond meticulously increase the prevalence of overall similarity classification.
Text mining in the classification of digital documents
Directory of Open Access Journals (Sweden)
Marcial Contreras Barrera
2016-11-01
Full Text Available Objective: Develop an automated classifier for the classification of bibliographic material by means of the text mining. Methodology: The text mining is used for the development of the classifier, based on a method of type supervised, conformed by two phases; learning and recognition, in the learning phase, the classifier learns patterns across the analysis of bibliographical records, of the classification Z, belonging to library science, information sciences and information resources, recovered from the database LIBRUNAM, in this phase is obtained the classifier capable of recognizing different subclasses (LC. In the recognition phase the classifier is validated and evaluates across classification tests, for this end bibliographical records of the classification Z are taken randomly, classified by a cataloguer and processed by the automated classifier, in order to obtain the precision of the automated classifier. Results: The application of the text mining achieved the development of the automated classifier, through the method classifying documents supervised type. The precision of the classifier was calculated doing the comparison among the assigned topics manually and automated obtaining 75.70% of precision. Conclusions: The application of text mining facilitated the creation of automated classifier, allowing to obtain useful technology for the classification of bibliographical material with the aim of improving and speed up the process of organizing digital documents.
Deep Multi-Task Learning for Tree Genera Classification
Ko, C.; Kang, J.; Sohn, G.
2018-05-01
The goal for our paper is to classify tree genera using airborne Light Detection and Ranging (LiDAR) data with Convolution Neural Network (CNN) - Multi-task Network (MTN) implementation. Unlike Single-task Network (STN) where only one task is assigned to the learning outcome, MTN is a deep learning architect for learning a main task (classification of tree genera) with other tasks (in our study, classification of coniferous and deciduous) simultaneously, with shared classification features. The main contribution of this paper is to improve classification accuracy from CNN-STN to CNN-MTN. This is achieved by introducing a concurrence loss (Lcd) to the designed MTN. This term regulates the overall network performance by minimizing the inconsistencies between the two tasks. Results show that we can increase the classification accuracy from 88.7 % to 91.0 % (from STN to MTN). The second goal of this paper is to solve the problem of small training sample size by multiple-view data generation. The motivation of this goal is to address one of the most common problems in implementing deep learning architecture, the insufficient number of training data. We address this problem by simulating training dataset with multiple-view approach. The promising results from this paper are providing a basis for classifying a larger number of dataset and number of classes in the future.
Directory of Open Access Journals (Sweden)
Zulma Liliana Sandoval Niño
2007-12-01
Full Text Available Se desarrolló un sistema de visión artificial para la clasificación de frutos de café en once categorías dependiendo de su estado de madurez. Para la descripción de la forma, el color y la textura de cada fruto de café se extrajeron 208 características. La reducción del conjunto de características de 208 a 9 se hizo con base en los resultados de dos métodos de selección de características, uno univariado y otro multivariado. Las características seleccionadas corresponden a 4 características de textura, 3 de color y 2 de forma. Este conjunto final de características se evaluó en dos técnicas de clasificación: Bayesiano y redes neuronales. Con el clasificador Bayesiano se obtuvo un error de clasificación del 5,43% y requirió un tiempo de clasificación de 5,5 ms, mientras que usando redes neuronales el error de clasificación fue de 7,46%, pero disminuyó el tiempo de clasificación a 0,8 ms.An artificial vision system for classification of coffee beans, in eleven categories, according to its state of maturity was developed. The description of the coffee beans was done by using 208 characteristics (form, color and texture characteristics. The reduction of the set of characteristics from 208 to 9 was done by using two methods of characteristic selection. The final set of characteristics is composed by 4 texture characteristics, 3 color characteristics and 2 shape characteristics. This final set was evaluated in two classifiers: The Bayesian and a neuronal networks classifier. The classification error obtained by the Bayesian classifier was 5,43%, it required 5,5 ms for the classification process, while the error obtained by neuronal networks classifier was 7,46% and the classification time decreased to 0,8 ms.
Coping with human errors through system design: Implications for ecological interface design
DEFF Research Database (Denmark)
Rasmussen, Jens; Vicente, Kim J.
1989-01-01
Research during recent years has revealed that human errors are not stochastic events which can be removed through improved training programs or optimal interface design. Rather, errors tend to reflect either systematic interference between various models, rules, and schemata, or the effects...... of the adaptive mechanisms involved in learning. In terms of design implications, these findings suggest that reliable human-system interaction will be achieved by designing interfaces which tend to minimize the potential for control interference and support recovery from errors. In other words, the focus should...... be on control of the effects of errors rather than on the elimination of errors per se. In this paper, we propose a theoretical framework for interface design that attempts to satisfy these objectives. The goal of our framework, called ecological interface design, is to develop a meaningful representation...
Sinha, S K; Karray, F
2002-01-01
Pipeline surface defects such as holes and cracks cause major problems for utility managers, particularly when the pipeline is buried under the ground. Manual inspection for surface defects in the pipeline has a number of drawbacks, including subjectivity, varying standards, and high costs. Automatic inspection system using image processing and artificial intelligence techniques can overcome many of these disadvantages and offer utility managers an opportunity to significantly improve quality and reduce costs. A recognition and classification of pipe cracks using images analysis and neuro-fuzzy algorithm is proposed. In the preprocessing step the scanned images of pipe are analyzed and crack features are extracted. In the classification step the neuro-fuzzy algorithm is developed that employs a fuzzy membership function and error backpropagation algorithm. The idea behind the proposed approach is that the fuzzy membership function will absorb variation of feature values and the backpropagation network, with its learning ability, will show good classification efficiency.
Hernández Farías, Delia Irazú; Laganà, Irene; Patti, Viviana; Bosco, Cristina
2018-01-01
The paper describes a preliminary study for the development of a novel lexicon for Italian sentiment analysis, i.e. where words are associated with polarity values. Given the influence of sentiment lexica on the performance of sentiment analysis systems, a methodology based on the detection and classification of errors in existing lexical resources is proposed and an extrinsic evaluation of the impact of such errors is applied. The final aim is to build a novel resource from the filtering app...
Statistical errors in Monte Carlo estimates of systematic errors
Energy Technology Data Exchange (ETDEWEB)
Roe, Byron P. [Department of Physics, University of Michigan, Ann Arbor, MI 48109 (United States)]. E-mail: byronroe@umich.edu
2007-01-01
For estimating the effects of a number of systematic errors on a data sample, one can generate Monte Carlo (MC) runs with systematic parameters varied and examine the change in the desired observed result. Two methods are often used. In the unisim method, the systematic parameters are varied one at a time by one standard deviation, each parameter corresponding to a MC run. In the multisim method (see ), each MC run has all of the parameters varied; the amount of variation is chosen from the expected distribution of each systematic parameter, usually assumed to be a normal distribution. The variance of the overall systematic error determination is derived for each of the two methods and comparisons are made between them. If one focuses not on the error in the prediction of an individual systematic error, but on the overall error due to all systematic errors in the error matrix element in data bin m, the number of events needed is strongly reduced because of the averaging effect over all of the errors. For simple models presented here the multisim model was far better if the statistical error in the MC samples was larger than an individual systematic error, while for the reverse case, the unisim model was better. Exact formulas and formulas for the simple toy models are presented so that realistic calculations can be made. The calculations in the present note are valid if the errors are in a linear region. If that region extends sufficiently far, one can have the unisims or multisims correspond to k standard deviations instead of one. This reduces the number of events required by a factor of k{sup 2}.
Statistical errors in Monte Carlo estimates of systematic errors
International Nuclear Information System (INIS)
Roe, Byron P.
2007-01-01
For estimating the effects of a number of systematic errors on a data sample, one can generate Monte Carlo (MC) runs with systematic parameters varied and examine the change in the desired observed result. Two methods are often used. In the unisim method, the systematic parameters are varied one at a time by one standard deviation, each parameter corresponding to a MC run. In the multisim method (see ), each MC run has all of the parameters varied; the amount of variation is chosen from the expected distribution of each systematic parameter, usually assumed to be a normal distribution. The variance of the overall systematic error determination is derived for each of the two methods and comparisons are made between them. If one focuses not on the error in the prediction of an individual systematic error, but on the overall error due to all systematic errors in the error matrix element in data bin m, the number of events needed is strongly reduced because of the averaging effect over all of the errors. For simple models presented here the multisim model was far better if the statistical error in the MC samples was larger than an individual systematic error, while for the reverse case, the unisim model was better. Exact formulas and formulas for the simple toy models are presented so that realistic calculations can be made. The calculations in the present note are valid if the errors are in a linear region. If that region extends sufficiently far, one can have the unisims or multisims correspond to k standard deviations instead of one. This reduces the number of events required by a factor of k 2
Determination of optimal samples for robot calibration based on error similarity
Directory of Open Access Journals (Sweden)
Tian Wei
2015-06-01
Full Text Available Industrial robots are used for automatic drilling and riveting. The absolute position accuracy of an industrial robot is one of the key performance indexes in aircraft assembly, and can be improved through error compensation to meet aircraft assembly requirements. The achievable accuracy and the difficulty of accuracy compensation implementation are closely related to the choice of sampling points. Therefore, based on the error similarity error compensation method, a method for choosing sampling points on a uniform grid is proposed. A simulation is conducted to analyze the influence of the sample point locations on error compensation. In addition, the grid steps of the sampling points are optimized using a statistical analysis method. The method is used to generate grids and optimize the grid steps of a Kuka KR-210 robot. The experimental results show that the method for planning sampling data can be used to effectively optimize the sampling grid. After error compensation, the position accuracy of the robot meets the position accuracy requirements.
Ringard, Justine; Seyler, Frederique; Linguet, Laurent
2017-06-16
Satellite precipitation products (SPPs) provide alternative precipitation data for regions with sparse rain gauge measurements. However, SPPs are subject to different types of error that need correction. Most SPP bias correction methods use the statistical properties of the rain gauge data to adjust the corresponding SPP data. The statistical adjustment does not make it possible to correct the pixels of SPP data for which there is no rain gauge data. The solution proposed in this article is to correct the daily SPP data for the Guiana Shield using a novel two set approach, without taking into account the daily gauge data of the pixel to be corrected, but the daily gauge data from surrounding pixels. In this case, a spatial analysis must be involved. The first step defines hydroclimatic areas using a spatial classification that considers precipitation data with the same temporal distributions. The second step uses the Quantile Mapping bias correction method to correct the daily SPP data contained within each hydroclimatic area. We validate the results by comparing the corrected SPP data and daily rain gauge measurements using relative RMSE and relative bias statistical errors. The results show that analysis scale variation reduces rBIAS and rRMSE significantly. The spatial classification avoids mixing rainfall data with different temporal characteristics in each hydroclimatic area, and the defined bias correction parameters are more realistic and appropriate. This study demonstrates that hydroclimatic classification is relevant for implementing bias correction methods at the local scale.
Trends and concepts in fern classification
Christenhusz, Maarten J. M.; Chase, Mark W.
2014-01-01
Background and Aims Throughout the history of fern classification, familial and generic concepts have been highly labile. Many classifications and evolutionary schemes have been proposed during the last two centuries, reflecting different interpretations of the available evidence. Knowledge of fern structure and life histories has increased through time, providing more evidence on which to base ideas of possible relationships, and classification has changed accordingly. This paper reviews previous classifications of ferns and presents ideas on how to achieve a more stable consensus. Scope An historical overview is provided from the first to the most recent fern classifications, from which conclusions are drawn on past changes and future trends. The problematic concept of family in ferns is discussed, with a particular focus on how this has changed over time. The history of molecular studies and the most recent findings are also presented. Key Results Fern classification generally shows a trend from highly artificial, based on an interpretation of a few extrinsic characters, via natural classifications derived from a multitude of intrinsic characters, towards more evolutionary circumscriptions of groups that do not in general align well with the distribution of these previously used characters. It also shows a progression from a few broad family concepts to systems that recognized many more narrowly and highly controversially circumscribed families; currently, the number of families recognized is stabilizing somewhere between these extremes. Placement of many genera was uncertain until the arrival of molecular phylogenetics, which has rapidly been improving our understanding of fern relationships. As a collective category, the so-called ‘fern allies’ (e.g. Lycopodiales, Psilotaceae, Equisetaceae) were unsurprisingly found to be polyphyletic, and the term should be abandoned. Lycopodiaceae, Selaginellaceae and Isoëtaceae form a clade (the lycopods) that is
Trends and concepts in fern classification.
Christenhusz, Maarten J M; Chase, Mark W
2014-03-01
Throughout the history of fern classification, familial and generic concepts have been highly labile. Many classifications and evolutionary schemes have been proposed during the last two centuries, reflecting different interpretations of the available evidence. Knowledge of fern structure and life histories has increased through time, providing more evidence on which to base ideas of possible relationships, and classification has changed accordingly. This paper reviews previous classifications of ferns and presents ideas on how to achieve a more stable consensus. An historical overview is provided from the first to the most recent fern classifications, from which conclusions are drawn on past changes and future trends. The problematic concept of family in ferns is discussed, with a particular focus on how this has changed over time. The history of molecular studies and the most recent findings are also presented. Fern classification generally shows a trend from highly artificial, based on an interpretation of a few extrinsic characters, via natural classifications derived from a multitude of intrinsic characters, towards more evolutionary circumscriptions of groups that do not in general align well with the distribution of these previously used characters. It also shows a progression from a few broad family concepts to systems that recognized many more narrowly and highly controversially circumscribed families; currently, the number of families recognized is stabilizing somewhere between these extremes. Placement of many genera was uncertain until the arrival of molecular phylogenetics, which has rapidly been improving our understanding of fern relationships. As a collective category, the so-called 'fern allies' (e.g. Lycopodiales, Psilotaceae, Equisetaceae) were unsurprisingly found to be polyphyletic, and the term should be abandoned. Lycopodiaceae, Selaginellaceae and Isoëtaceae form a clade (the lycopods) that is sister to all other vascular plants, whereas
Achieving excellence on shift through teamwork
International Nuclear Information System (INIS)
Newman, L.
1988-01-01
Anyone familiar with the nuclear industry realizes the importance of operators. Operators can achieve error-free plant operations, i.e., excellence on shift through teamwork. As a shift supervisor (senior reactor operator/shift technical advisor) the author went through the plant's first cycle of operations with no scrams and no equipment damaged by operator error, having since changed roles (and companies) to one of assessing plant operations. This change has provided the opportunity to see objectively the importance of operators working together and of the team building and teamwork that contribute to the shift's success. This paper uses examples to show the effectiveness of working together and outlines steps for building a group of operators into a team
A precise error bound for quantum phase estimation.
Directory of Open Access Journals (Sweden)
James M Chappell
Full Text Available Quantum phase estimation is one of the key algorithms in the field of quantum computing, but up until now, only approximate expressions have been derived for the probability of error. We revisit these derivations, and find that by ensuring symmetry in the error definitions, an exact formula can be found. This new approach may also have value in solving other related problems in quantum computing, where an expected error is calculated. Expressions for two special cases of the formula are also developed, in the limit as the number of qubits in the quantum computer approaches infinity and in the limit as the extra added qubits to improve reliability goes to infinity. It is found that this formula is useful in validating computer simulations of the phase estimation procedure and in avoiding the overestimation of the number of qubits required in order to achieve a given reliability. This formula thus brings improved precision in the design of quantum computers.
Classification Based on Pruning and Double Covered Rule Sets for the Internet of Things Applications
Zhou, Zhongmei; Wang, Weiping
2014-01-01
The Internet of things (IOT) is a hot issue in recent years. It accumulates large amounts of data by IOT users, which is a great challenge to mining useful knowledge from IOT. Classification is an effective strategy which can predict the need of users in IOT. However, many traditional rule-based classifiers cannot guarantee that all instances can be covered by at least two classification rules. Thus, these algorithms cannot achieve high accuracy in some datasets. In this paper, we propose a new rule-based classification, CDCR-P (Classification based on the Pruning and Double Covered Rule sets). CDCR-P can induce two different rule sets A and B. Every instance in training set can be covered by at least one rule not only in rule set A, but also in rule set B. In order to improve the quality of rule set B, we take measure to prune the length of rules in rule set B. Our experimental results indicate that, CDCR-P not only is feasible, but also it can achieve high accuracy. PMID:24511304
Li, Shasha; Zhou, Zhongmei; Wang, Weiping
2014-01-01
The Internet of things (IOT) is a hot issue in recent years. It accumulates large amounts of data by IOT users, which is a great challenge to mining useful knowledge from IOT. Classification is an effective strategy which can predict the need of users in IOT. However, many traditional rule-based classifiers cannot guarantee that all instances can be covered by at least two classification rules. Thus, these algorithms cannot achieve high accuracy in some datasets. In this paper, we propose a new rule-based classification, CDCR-P (Classification based on the Pruning and Double Covered Rule sets). CDCR-P can induce two different rule sets A and B. Every instance in training set can be covered by at least one rule not only in rule set A, but also in rule set B. In order to improve the quality of rule set B, we take measure to prune the length of rules in rule set B. Our experimental results indicate that, CDCR-P not only is feasible, but also it can achieve high accuracy.
2016-06-10
use different terminology depending on which sister service they are from. Every service has various medical capabilities for each role of medical ... Medical Errors, Combat Casualty Care, Culture of Safety 16. SECURITY CLASSIFICATION OF: 17. LIMITATION OF ABSTRACT 18. NUMBER OF PAGES 19a...Army) AE Adverse event AHRQ Agency for Healthcare Research and Quality AHS Army Health System AMEDD Army Medical Department CPQ Clinical Practice
An Efficient Ensemble Learning Method for Gene Microarray Classification
Directory of Open Access Journals (Sweden)
Alireza Osareh
2013-01-01
Full Text Available The gene microarray analysis and classification have demonstrated an effective way for the effective diagnosis of diseases and cancers. However, it has been also revealed that the basic classification techniques have intrinsic drawbacks in achieving accurate gene classification and cancer diagnosis. On the other hand, classifier ensembles have received increasing attention in various applications. Here, we address the gene classification issue using RotBoost ensemble methodology. This method is a combination of Rotation Forest and AdaBoost techniques which in turn preserve both desirable features of an ensemble architecture, that is, accuracy and diversity. To select a concise subset of informative genes, 5 different feature selection algorithms are considered. To assess the efficiency of the RotBoost, other nonensemble/ensemble techniques including Decision Trees, Support Vector Machines, Rotation Forest, AdaBoost, and Bagging are also deployed. Experimental results have revealed that the combination of the fast correlation-based feature selection method with ICA-based RotBoost ensemble is highly effective for gene classification. In fact, the proposed method can create ensemble classifiers which outperform not only the classifiers produced by the conventional machine learning but also the classifiers generated by two widely used conventional ensemble learning methods, that is, Bagging and AdaBoost.
Energy Technology Data Exchange (ETDEWEB)
Elliott, C.J.; McVey, B. (Los Alamos National Lab., NM (USA)); Quimby, D.C. (Spectra Technology, Inc., Bellevue, WA (USA))
1990-01-01
The level of field errors in an FEL is an important determinant of its performance. We have computed 3D performance of a large laser subsystem subjected to field errors of various types. These calculations have been guided by simple models such as SWOOP. The technique of choice is utilization of the FELEX free electron laser code that now possesses extensive engineering capabilities. Modeling includes the ability to establish tolerances of various types: fast and slow scale field bowing, field error level, beam position monitor error level, gap errors, defocusing errors, energy slew, displacement and pointing errors. Many effects of these errors on relative gain and relative power extraction are displayed and are the essential elements of determining an error budget. The random errors also depend on the particular random number seed used in the calculation. The simultaneous display of the performance versus error level of cases with multiple seeds illustrates the variations attributable to stochasticity of this model. All these errors are evaluated numerically for comprehensive engineering of the system. In particular, gap errors are found to place requirements beyond mechanical tolerances of {plus minus}25{mu}m, and amelioration of these may occur by a procedure utilizing direct measurement of the magnetic fields at assembly time. 4 refs., 12 figs.
A Comprehensive Study of Features and Algorithms for URL-Based Topic Classification
Weber, I; Henzinger, M; Baykan, E
2011-01-01
Given only the URL of a Web page, can we identify its topic? We study this problem in detail by exploring a large number of different feature sets and algorithms on several datasets. We also show that the inherent overlap between topics and the sparsity of the information in URLs makes this a very challenging problem. Web page classification without a page's content is desirable when the content is not available at all, when a classification is needed before obtaining the content, or when classification speed is of utmost importance. For our experiments we used five different corpora comprising a total of about 3 million (URL, classification) pairs. We evaluated several techniques for feature generation and classification algorithms. The individual binary classifiers were then combined via boosting into metabinary classifiers. We achieve typical F-measure values between 80 and 85, and a typical precision of around 86. The precision can be pushed further over 90 while maintaining a typical level of recall betw...
Qi, K.; Qingfeng, G.
2017-12-01
With the popular use of High-Resolution Satellite (HRS) images, more and more research efforts have been placed on land-use scene classification. However, it makes the task difficult with HRS images for the complex background and multiple land-cover classes or objects. This article presents a multiscale deeply described correlaton model for land-use scene classification. Specifically, the convolutional neural network is introduced to learn and characterize the local features at different scales. Then, learnt multiscale deep features are explored to generate visual words. The spatial arrangement of visual words is achieved through the introduction of adaptive vector quantized correlograms at different scales. Experiments on two publicly available land-use scene datasets demonstrate that the proposed model is compact and yet discriminative for efficient representation of land-use scene images, and achieves competitive classification results with the state-of-art methods.
Bryant, C. M.; Prudhomme, S.; Wildey, T.
2015-01-01
In this work, we investigate adaptive approaches to control errors in response surface approximations computed from numerical approximations of differential equations with uncertain or random data and coefficients. The adaptivity of the response surface approximation is based on a posteriori error estimation, and the approach relies on the ability to decompose the a posteriori error estimate into contributions from the physical discretization and the approximation in parameter space. Errors are evaluated in terms of linear quantities of interest using adjoint-based methodologies. We demonstrate that a significant reduction in the computational cost required to reach a given error tolerance can be achieved by refining the dominant error contributions rather than uniformly refining both the physical and stochastic discretization. Error decomposition is demonstrated for a two-dimensional flow problem, and adaptive procedures are tested on a convection-diffusion problem with discontinuous parameter dependence and a diffusion problem, where the diffusion coefficient is characterized by a 10-dimensional parameter space.
Belgiu, Mariana; ǎguţ, Lucian, , Dr; Strobl, Josef
2014-01-01
The increasing availability of high resolution imagery has triggered the need for automated image analysis techniques, with reduced human intervention and reproducible analysis procedures. The knowledge gained in the past might be of use to achieving this goal, if systematically organized into libraries which would guide the image analysis procedure. In this study we aimed at evaluating the variability of digital classifications carried out by three experts who were all assigned the same interpretation task. Besides the three classifications performed by independent operators, we developed an additional rule-based classification that relied on the image classifications best practices found in the literature, and used it as a surrogate for libraries of object characteristics. The results showed statistically significant differences among all operators who classified the same reference imagery. The classifications carried out by the experts achieved satisfactory results when transferred to another area for extracting the same classes of interest, without modification of the developed rules.
CLASS-PAIR-GUIDED MULTIPLE KERNEL LEARNING OF INTEGRATING HETEROGENEOUS FEATURES FOR CLASSIFICATION
Directory of Open Access Journals (Sweden)
Q. Wang
2017-10-01
Full Text Available In recent years, many studies on remote sensing image classification have shown that using multiple features from different data sources can effectively improve the classification accuracy. As a very powerful means of learning, multiple kernel learning (MKL can conveniently be embedded in a variety of characteristics. The conventional combined kernel learned by MKL can be regarded as the compromise of all basic kernels for all classes in classification. It is the best of the whole, but not optimal for each specific class. For this problem, this paper proposes a class-pair-guided MKL method to integrate the heterogeneous features (HFs from multispectral image (MSI and light detection and ranging (LiDAR data. In particular, the one-against-one strategy is adopted, which converts multiclass classification problem to a plurality of two-class classification problem. Then, we select the best kernel from pre-constructed basic kernels set for each class-pair by kernel alignment (KA in the process of classification. The advantage of the proposed method is that only the best kernel for the classification of any two classes can be retained, which leads to greatly enhanced discriminability. Experiments are conducted on two real data sets, and the experimental results show that the proposed method achieves the best performance in terms of classification accuracies in integrating the HFs for classification when compared with several state-of-the-art algorithms.
Cole, Russell; Haimson, Joshua; Perez-Johnson, Irma; May, Henry
2011-01-01
State assessments are increasingly used as outcome measures for education evaluations. The scaling of state assessments produces variability in measurement error, with the conditional standard error of measurement increasing as average student ability moves toward the tails of the achievement distribution. This report examines the variability in…
A continuous quality improvement project to reduce medication error in the emergency department.
Lee, Sara Bc; Lee, Larry Ly; Yeung, Richard Sd; Chan, Jimmy Ts
2013-01-01
Medication errors are a common source of adverse healthcare incidents particularly in the emergency department (ED) that has a number of factors that make it prone to medication errors. This project aims to reduce medication errors and improve the health and economic outcomes of clinical care in Hong Kong ED. In 2009, a task group was formed to identify problems that potentially endanger medication safety and developed strategies to eliminate these problems. Responsible officers were assigned to look after seven error-prone areas. Strategies were proposed, discussed, endorsed and promulgated to eliminate the problems identified. A reduction of medication incidents (MI) from 16 to 6 was achieved before and after the improvement work. This project successfully established a concrete organizational structure to safeguard error-prone areas of medication safety in a sustainable manner.
Deep Recurrent Neural Networks for Supernovae Classification
Charnock, Tom; Moss, Adam
2017-03-01
We apply deep recurrent neural networks, which are capable of learning complex sequential information, to classify supernovae (code available at https://github.com/adammoss/supernovae). The observational time and filter fluxes are used as inputs to the network, but since the inputs are agnostic, additional data such as host galaxy information can also be included. Using the Supernovae Photometric Classification Challenge (SPCC) data, we find that deep networks are capable of learning about light curves, however the performance of the network is highly sensitive to the amount of training data. For a training size of 50% of the representational SPCC data set (around 104 supernovae) we obtain a type-Ia versus non-type-Ia classification accuracy of 94.7%, an area under the Receiver Operating Characteristic curve AUC of 0.986 and an SPCC figure-of-merit F 1 = 0.64. When using only the data for the early-epoch challenge defined by the SPCC, we achieve a classification accuracy of 93.1%, AUC of 0.977, and F 1 = 0.58, results almost as good as with the whole light curve. By employing bidirectional neural networks, we can acquire impressive classification results between supernovae types I, II and III at an accuracy of 90.4% and AUC of 0.974. We also apply a pre-trained model to obtain classification probabilities as a function of time and show that it can give early indications of supernovae type. Our method is competitive with existing algorithms and has applications for future large-scale photometric surveys.
Water displacement leg volumetry in clinical studies - A discussion of error sources
2010-01-01
Background Water displacement leg volumetry is a highly reproducible method, allowing the confirmation of efficacy of vasoactive substances. Nevertheless errors of its execution and the selection of unsuitable patients are likely to negatively affect the outcome of clinical studies in chronic venous insufficiency (CVI). Discussion Placebo controlled double-blind drug studies in CVI were searched (Cochrane Review 2005, MedLine Search until December 2007) and assessed with regard to efficacy (volume reduction of the leg), patient characteristics, and potential methodological error sources. Almost every second study reported only small drug effects (≤ 30 mL volume reduction). As the most relevant error source the conduct of volumetry was identified. Because the practical use of available equipment varies, volume differences of more than 300 mL - which is a multifold of a potential treatment effect - have been reported between consecutive measurements. Other potential error sources were insufficient patient guidance or difficulties with the transition from the Widmer CVI classification to the CEAP (Clinical Etiological Anatomical Pathophysiological) grading. Summary Patients should be properly diagnosed with CVI and selected for stable oedema and further clinical symptoms relevant for the specific study. Centres require a thorough training on the use of the volumeter and on patient guidance. Volumetry should be performed under constant conditions. The reproducibility of short term repeat measurements has to be ensured. PMID:20070899
ERF/ERFC, Calculation of Error Function, Complementary Error Function, Probability Integrals
International Nuclear Information System (INIS)
Vogel, J.E.
1983-01-01
1 - Description of problem or function: ERF and ERFC are used to compute values of the error function and complementary error function for any real number. They may be used to compute other related functions such as the normal probability integrals. 4. Method of solution: The error function and complementary error function are approximated by rational functions. Three such rational approximations are used depending on whether - x .GE.4.0. In the first region the error function is computed directly and the complementary error function is computed via the identity erfc(x)=1.0-erf(x). In the other two regions the complementary error function is computed directly and the error function is computed from the identity erf(x)=1.0-erfc(x). The error function and complementary error function are real-valued functions of any real argument. The range of the error function is (-1,1). The range of the complementary error function is (0,2). 5. Restrictions on the complexity of the problem: The user is cautioned against using ERF to compute the complementary error function by using the identity erfc(x)=1.0-erf(x). This subtraction may cause partial or total loss of significance for certain values of x
SAW Classification Algorithm for Chinese Text Classification
Xiaoli Guo; Huiyu Sun; Tiehua Zhou; Ling Wang; Zhaoyang Qu; Jiannan Zang
2015-01-01
Considering the explosive growth of data, the increased amount of text data’s effect on the performance of text categorization forward the need for higher requirements, such that the existing classification method cannot be satisfied. Based on the study of existing text classification technology and semantics, this paper puts forward a kind of Chinese text classification oriented SAW (Structural Auxiliary Word) algorithm. The algorithm uses the special space effect of Chinese text where words...
Achieving few-femtosecond time-sorting at hard X-ray free-electron lasers
Harmand, M.; Coffee, R.; Bionta, M. R.; Chollet, M.; French, D.; Zhu, D.; Fritz, D. M.; Lemke, H. T.; Medvedev, N.; Ziaja, B.; Toleikis, S.; Cammarata, M.
2013-03-01
Recently, few-femtosecond pulses have become available at hard X-ray free-electron lasers. Coupled with the available sub-10 fs optical pulses, investigations into few-femtosecond dynamics are not far off. However, achieving sufficient synchronization between optical lasers and X-ray pulses continues to be challenging. We report a `measure-and-sort' approach, which achieves sub-10 fs root-mean-squared (r.m.s.) error measurement at hard X-ray FELs, far beyond the 100-200 fs r.m.s. jitter limitations. This timing diagnostic, now routinely available at the Linac Coherent Light Source (LCLS), is based on ultrafast free-carrier generation in optically transparent materials. Correlation between two independent measurements enables unambiguous demonstration of ~6 fs r.m.s. error in reporting the optical/X-ray delay, with single shot error suggesting the possibility of reaching few-femtosecond resolution.
DEFF Research Database (Denmark)
Mai, Jens Erik
2004-01-01
This paper surveys classification research literature, discusses various classification theories, and shows that the focus has traditionally been on establishing a scientific foundation for classification research. This paper argues that a shift has taken place, and suggests that contemporary...... classification research focus on contextual information as the guide for the design and construction of classification schemes....
Classification and data acquisition with incomplete data
Williams, David P.
In remote-sensing applications, incomplete data can result when only a subset of sensors (e.g., radar, infrared, acoustic) are deployed at certain regions. The limitations of single sensor systems have spurred interest in employing multiple sensor modalities simultaneously. For example, in land mine detection tasks, different sensor modalities are better-suited to capture different aspects of the underlying physics of the mines. Synthetic aperture radar sensors may be better at detecting surface mines, while infrared sensors may be better at detecting buried mines. By employing multiple sensor modalities to address the detection task, the strengths of the disparate sensors can be exploited in a synergistic manner to improve performance beyond that which would be achievable with either single sensor alone. When multi-sensor approaches are employed, however, incomplete data can be manifested. If each sensor is located on a separate platform ( e.g., aircraft), each sensor may interrogate---and hence collect data over---only partially overlapping areas of land. As a result, some data points may be characterized by data (i.e., features) from only a subset of the possible sensors employed in the task. Equivalently, this scenario implies that some data points will be missing features. Increasing focus in the future on using---and fusing data from---multiple sensors will make such incomplete-data problems commonplace. In many applications involving incomplete data, it is possible to acquire the missing data at a cost. In multi-sensor remote-sensing applications, data is acquired by deploying sensors to data points. Acquiring data is usually an expensive, time-consuming task, a fact that necessitates an intelligent data acquisition process. Incomplete data is not limited to remote-sensing applications, but rather, can arise in virtually any data set. In this dissertation, we address the general problem of classification when faced with incomplete data. We also address the
Classification of amyloid status using machine learning with histograms of oriented 3D gradients
Directory of Open Access Journals (Sweden)
Liam Cattell
2016-01-01
Full Text Available Brain amyloid burden may be quantitatively assessed from positron emission tomography imaging using standardised uptake value ratios. Using these ratios as an adjunct to visual image assessment has been shown to improve inter-reader reliability, however, the amyloid positivity threshold is dependent on the tracer and specific image regions used to calculate the uptake ratio. To address this problem, we propose a machine learning approach to amyloid status classification, which is independent of tracer and does not require a specific set of regions of interest. Our method extracts feature vectors from amyloid images, which are based on histograms of oriented three-dimensional gradients. We optimised our method on 133 18F-florbetapir brain volumes, and applied it to a separate test set of 131 volumes. Using the same parameter settings, we then applied our method to 209 11C-PiB images and 128 18F-florbetaben images. We compared our method to classification results achieved using two other methods: standardised uptake value ratios and a machine learning method based on voxel intensities. Our method resulted in the largest mean distances between the subjects and the classification boundary, suggesting that it is less likely to make low-confidence classification decisions. Moreover, our method obtained the highest classification accuracy for all three tracers, and consistently achieved above 96% accuracy.
Yohay Carmel; Curtis Flather; Denis Dean
2006-01-01
This paper summarizes our efforts to investigate the nature, behavior, and implications of positional error and attribute error in spatiotemporal datasets. Estimating the combined influence of these errors on map analysis has been hindered by the fact that these two error types are traditionally expressed in different units (distance units, and categorical units,...
Error minimizing algorithms for nearest eighbor classifiers
Energy Technology Data Exchange (ETDEWEB)
Porter, Reid B [Los Alamos National Laboratory; Hush, Don [Los Alamos National Laboratory; Zimmer, G. Beate [TEXAS A& M
2011-01-03
Stack Filters define a large class of discrete nonlinear filter first introd uced in image and signal processing for noise removal. In recent years we have suggested their application to classification problems, and investigated their relationship to other types of discrete classifiers such as Decision Trees. In this paper we focus on a continuous domain version of Stack Filter Classifiers which we call Ordered Hypothesis Machines (OHM), and investigate their relationship to Nearest Neighbor classifiers. We show that OHM classifiers provide a novel framework in which to train Nearest Neighbor type classifiers by minimizing empirical error based loss functions. We use the framework to investigate a new cost sensitive loss function that allows us to train a Nearest Neighbor type classifier for low false alarm rate applications. We report results on both synthetic data and real-world image data.
DEFF Research Database (Denmark)
Hjørland, Birger
2017-01-01
This article presents and discusses definitions of the term “classification” and the related concepts “Concept/conceptualization,”“categorization,” “ordering,” “taxonomy” and “typology.” It further presents and discusses theories of classification including the influences of Aristotle...... and Wittgenstein. It presents different views on forming classes, including logical division, numerical taxonomy, historical classification, hermeneutical and pragmatic/critical views. Finally, issues related to artificial versus natural classification and taxonomic monism versus taxonomic pluralism are briefly...
Classification of ECG beats using deep belief network and active learning.
G, Sayantan; T, Kien P; V, Kadambari K
2018-04-12
A new semi-supervised approach based on deep learning and active learning for classification of electrocardiogram signals (ECG) is proposed. The objective of the proposed work is to model a scientific method for classification of cardiac irregularities using electrocardiogram beats. The model follows the Association for the Advancement of medical instrumentation (AAMI) standards and consists of three phases. In phase I, feature representation of ECG is learnt using Gaussian-Bernoulli deep belief network followed by a linear support vector machine (SVM) training in the consecutive phase. It yields three deep models which are based on AAMI-defined classes, namely N, V, S, and F. In the last phase, a query generator is introduced to interact with the expert to label few beats to improve accuracy and sensitivity. The proposed approach depicts significant improvement in accuracy with minimal queries posed to the expert and fast online training as tested on the MIT-BIH Arrhythmia Database and the MIT-BIH Supra-ventricular Arrhythmia Database (SVDB). With 100 queries labeled by the expert in phase III, the method achieves an accuracy of 99.5% in "S" versus all classifications (SVEB) and 99.4% accuracy in "V " versus all classifications (VEB) on MIT-BIH Arrhythmia Database. In a similar manner, it is attributed that an accuracy of 97.5% for SVEB and 98.6% for VEB on SVDB database is achieved respectively. Graphical Abstract Reply- Deep belief network augmented by active learning for efficient prediction of arrhythmia.
Chen, Chau-Kuang
2010-01-01
Artificial Neural Network (ANN) and Support Vector Machine (SVM) approaches have been on the cutting edge of science and technology for pattern recognition and data classification. In the ANN model, classification accuracy can be achieved by using the feed-forward of inputs, back-propagation of errors, and the adjustment of connection weights. In…
International Nuclear Information System (INIS)
Velez R, Luis Anibal; Largacha R, Antonio
2003-01-01
Regulations concerning land use density controls in non-urban areas are often ineffective in protecting the rural character of such areas, within the context of Colombia's territorial organization plans What they tend to do is indirectly promote urban expansion through the fixing of minimum plot sizes Rather than question the methods employed in land use classification, the present study uses the case of the rural area of Santa Elena on the outskirts of Medellin and the existing zoning controls as established in the territorial organization plan, and focuses on the regulations established for each land use category as they effect occupation densities and patterns of plot fragmentation. On the one hand a simulation is undertaken of the trends which would result from land occupation in accordance with existing regulations concerning land use classification and plot size This indicates that the overall effect would be to disperse settlement patterns and fragment the landscape Secondly, an alternative scenario is developed based on the modification of minimum plot sizes for each of the three land use classifications established in the existing plan (protection, rural and suburban) In this way, and through the perspective of landscape ecology, It Is shown that in certain cases less dispersion and greater concentration of settlements can be achieved, and in other cases dispersion is minimized The use of GIS is fundamental in the development of such simulation techniques
Li, Mengmeng; Bijker, Wietske; Stein, Alfred
2015-04-01
Two main challenges are faced when classifying urban land cover from very high resolution satellite images: obtaining an optimal image segmentation and distinguishing buildings from other man-made objects. For optimal segmentation, this work proposes a hierarchical representation of an image by means of a Binary Partition Tree (BPT) and an unsupervised evaluation of image segmentations by energy minimization. For building extraction, we apply fuzzy sets to create a fuzzy landscape of shadows which in turn involves a two-step procedure. The first step is a preliminarily image classification at a fine segmentation level to generate vegetation and shadow information. The second step models the directional relationship between building and shadow objects to extract building information at the optimal segmentation level. We conducted the experiments on two datasets of Pléiades images from Wuhan City, China. To demonstrate its performance, the proposed classification is compared at the optimal segmentation level with Maximum Likelihood Classification and Support Vector Machine classification. The results show that the proposed classification produced the highest overall accuracies and kappa coefficients, and the smallest over-classification and under-classification geometric errors. We conclude first that integrating BPT with energy minimization offers an effective means for image segmentation. Second, we conclude that the directional relationship between building and shadow objects represented by a fuzzy landscape is important for building extraction.
A new classification scheme of plastic wastes based upon recycling labels
Energy Technology Data Exchange (ETDEWEB)
Özkan, Kemal, E-mail: kozkan@ogu.edu.tr [Computer Engineering Dept., Eskişehir Osmangazi University, 26480 Eskişehir (Turkey); Ergin, Semih, E-mail: sergin@ogu.edu.tr [Electrical Electronics Engineering Dept., Eskişehir Osmangazi University, 26480 Eskişehir (Turkey); Işık, Şahin, E-mail: sahini@ogu.edu.tr [Computer Engineering Dept., Eskişehir Osmangazi University, 26480 Eskişehir (Turkey); Işıklı, İdil, E-mail: idil.isikli@bilecik.edu.tr [Electrical Electronics Engineering Dept., Bilecik University, 11210 Bilecik (Turkey)
2015-01-15
Highlights: • PET, HPDE or PP types of plastics are considered. • An automated classification of plastic bottles based on the feature extraction and classification methods is performed. • The decision mechanism consists of PCA, Kernel PCA, FLDA, SVD and Laplacian Eigenmaps methods. • SVM is selected to achieve the classification task and majority voting technique is used. - Abstract: Since recycling of materials is widely assumed to be environmentally and economically beneficial, reliable sorting and processing of waste packaging materials such as plastics is very important for recycling with high efficiency. An automated system that can quickly categorize these materials is certainly needed for obtaining maximum classification while maintaining high throughput. In this paper, first of all, the photographs of the plastic bottles have been taken and several preprocessing steps were carried out. The first preprocessing step is to extract the plastic area of a bottle from the background. Then, the morphological image operations are implemented. These operations are edge detection, noise removal, hole removing, image enhancement, and image segmentation. These morphological operations can be generally defined in terms of the combinations of erosion and dilation. The effect of bottle color as well as label are eliminated using these operations. Secondly, the pixel-wise intensity values of the plastic bottle images have been used together with the most popular subspace and statistical feature extraction methods to construct the feature vectors in this study. Only three types of plastics are considered due to higher existence ratio of them than the other plastic types in the world. The decision mechanism consists of five different feature extraction methods including as Principal Component Analysis (PCA), Kernel PCA (KPCA), Fisher’s Linear Discriminant Analysis (FLDA), Singular Value Decomposition (SVD) and Laplacian Eigenmaps (LEMAP) and uses a simple
A new classification scheme of plastic wastes based upon recycling labels
International Nuclear Information System (INIS)
Özkan, Kemal; Ergin, Semih; Işık, Şahin; Işıklı, İdil
2015-01-01
Highlights: • PET, HPDE or PP types of plastics are considered. • An automated classification of plastic bottles based on the feature extraction and classification methods is performed. • The decision mechanism consists of PCA, Kernel PCA, FLDA, SVD and Laplacian Eigenmaps methods. • SVM is selected to achieve the classification task and majority voting technique is used. - Abstract: Since recycling of materials is widely assumed to be environmentally and economically beneficial, reliable sorting and processing of waste packaging materials such as plastics is very important for recycling with high efficiency. An automated system that can quickly categorize these materials is certainly needed for obtaining maximum classification while maintaining high throughput. In this paper, first of all, the photographs of the plastic bottles have been taken and several preprocessing steps were carried out. The first preprocessing step is to extract the plastic area of a bottle from the background. Then, the morphological image operations are implemented. These operations are edge detection, noise removal, hole removing, image enhancement, and image segmentation. These morphological operations can be generally defined in terms of the combinations of erosion and dilation. The effect of bottle color as well as label are eliminated using these operations. Secondly, the pixel-wise intensity values of the plastic bottle images have been used together with the most popular subspace and statistical feature extraction methods to construct the feature vectors in this study. Only three types of plastics are considered due to higher existence ratio of them than the other plastic types in the world. The decision mechanism consists of five different feature extraction methods including as Principal Component Analysis (PCA), Kernel PCA (KPCA), Fisher’s Linear Discriminant Analysis (FLDA), Singular Value Decomposition (SVD) and Laplacian Eigenmaps (LEMAP) and uses a simple
Statistical errors in Monte Carlo estimates of systematic errors
Roe, Byron P.
2007-01-01
For estimating the effects of a number of systematic errors on a data sample, one can generate Monte Carlo (MC) runs with systematic parameters varied and examine the change in the desired observed result. Two methods are often used. In the unisim method, the systematic parameters are varied one at a time by one standard deviation, each parameter corresponding to a MC run. In the multisim method (see ), each MC run has all of the parameters varied; the amount of variation is chosen from the expected distribution of each systematic parameter, usually assumed to be a normal distribution. The variance of the overall systematic error determination is derived for each of the two methods and comparisons are made between them. If one focuses not on the error in the prediction of an individual systematic error, but on the overall error due to all systematic errors in the error matrix element in data bin m, the number of events needed is strongly reduced because of the averaging effect over all of the errors. For simple models presented here the multisim model was far better if the statistical error in the MC samples was larger than an individual systematic error, while for the reverse case, the unisim model was better. Exact formulas and formulas for the simple toy models are presented so that realistic calculations can be made. The calculations in the present note are valid if the errors are in a linear region. If that region extends sufficiently far, one can have the unisims or multisims correspond to k standard deviations instead of one. This reduces the number of events required by a factor of k2. The specific terms unisim and multisim were coined by Peter Meyers and Steve Brice, respectively, for the MiniBooNE experiment. However, the concepts have been developed over time and have been in general use for some time.
Addressing the Problem of Negative Lexical Transfer Errors in Chilean University Students
Directory of Open Access Journals (Sweden)
Paul Anthony Dissington
2018-01-01
Full Text Available Studies of second language learning have revealed a connection between first language transfer and errors in second language production. This paper describes an action research study carried out among Chilean university students studying English as part of their degree programmes. The study focuses on common lexical errors made by Chilean Spanish-speakers due to negative first language transfer and aims to analyse the effects of systematic instruction and practice of this problematic lexis. It is suggested that raising awareness of lexical transfer through focused attention on common transfer errors is valued by students and seems essential for learners to achieve productive mastery.
Tao, C.-S.; Chen, S.-W.; Li, Y.-Z.; Xiao, S.-P.
2017-09-01
Land cover classification is an important application for polarimetric synthetic aperture radar (PolSAR) data utilization. Rollinvariant polarimetric features such as H / Ani / text-decoration: overline">α / Span are commonly adopted in PolSAR land cover classification. However, target orientation diversity effect makes PolSAR images understanding and interpretation difficult. Only using the roll-invariant polarimetric features may introduce ambiguity in the interpretation of targets' scattering mechanisms and limit the followed classification accuracy. To address this problem, this work firstly focuses on hidden polarimetric feature mining in the rotation domain along the radar line of sight using the recently reported uniform polarimetric matrix rotation theory and the visualization and characterization tool of polarimetric coherence pattern. The former rotates the acquired polarimetric matrix along the radar line of sight and fully describes the rotation characteristics of each entry of the matrix. Sets of new polarimetric features are derived to describe the hidden scattering information of the target in the rotation domain. The latter extends the traditional polarimetric coherence at a given rotation angle to the rotation domain for complete interpretation. A visualization and characterization tool is established to derive new polarimetric features for hidden information exploration. Then, a classification scheme is developed combing both the selected new hidden polarimetric features in rotation domain and the commonly used roll-invariant polarimetric features with a support vector machine (SVM) classifier. Comparison experiments based on AIRSAR and multi-temporal UAVSAR data demonstrate that compared with the conventional classification scheme which only uses the roll-invariant polarimetric features, the proposed classification scheme achieves both higher classification accuracy and better robustness. For AIRSAR data, the overall classification
Directory of Open Access Journals (Sweden)
C.-S. Tao
2017-09-01
Full Text Available Land cover classification is an important application for polarimetric synthetic aperture radar (PolSAR data utilization. Rollinvariant polarimetric features such as H / Ani / α / Span are commonly adopted in PolSAR land cover classification. However, target orientation diversity effect makes PolSAR images understanding and interpretation difficult. Only using the roll-invariant polarimetric features may introduce ambiguity in the interpretation of targets’ scattering mechanisms and limit the followed classification accuracy. To address this problem, this work firstly focuses on hidden polarimetric feature mining in the rotation domain along the radar line of sight using the recently reported uniform polarimetric matrix rotation theory and the visualization and characterization tool of polarimetric coherence pattern. The former rotates the acquired polarimetric matrix along the radar line of sight and fully describes the rotation characteristics of each entry of the matrix. Sets of new polarimetric features are derived to describe the hidden scattering information of the target in the rotation domain. The latter extends the traditional polarimetric coherence at a given rotation angle to the rotation domain for complete interpretation. A visualization and characterization tool is established to derive new polarimetric features for hidden information exploration. Then, a classification scheme is developed combing both the selected new hidden polarimetric features in rotation domain and the commonly used roll-invariant polarimetric features with a support vector machine (SVM classifier. Comparison experiments based on AIRSAR and multi-temporal UAVSAR data demonstrate that compared with the conventional classification scheme which only uses the roll-invariant polarimetric features, the proposed classification scheme achieves both higher classification accuracy and better robustness. For AIRSAR data, the overall classification accuracy
Thin film thickness measurement error reduction by wavelength selection in spectrophotometry
International Nuclear Information System (INIS)
Tsepulin, Vladimir G; Perchik, Alexey V; Tolstoguzov, Victor L; Karasik, Valeriy E
2015-01-01
Fast and accurate volumetric profilometry of thin film structures is an important problem in the electronic visual display industry. We propose to use spectrophotometry with a limited number of working wavelengths to achieve high-speed control and an approach to selecting the optimal working wavelengths to reduce the thickness measurement error. A simple expression for error estimation is presented and tested using a Monte Carlo simulation. The experimental setup is designed to confirm the stability of film thickness determination using a limited number of wavelengths
An object-oriented classification method of high resolution imagery based on improved AdaTree
International Nuclear Information System (INIS)
Xiaohe, Zhang; Liang, Zhai; Jixian, Zhang; Huiyong, Sang
2014-01-01
With the popularity of the application using high spatial resolution remote sensing image, more and more studies paid attention to object-oriented classification on image segmentation as well as automatic classification after image segmentation. This paper proposed a fast method of object-oriented automatic classification. First, edge-based or FNEA-based segmentation was used to identify image objects and the values of most suitable attributes of image objects for classification were calculated. Then a certain number of samples from the image objects were selected as training data for improved AdaTree algorithm to get classification rules. Finally, the image objects could be classified easily using these rules. In the AdaTree, we mainly modified the final hypothesis to get classification rules. In the experiment with WorldView2 image, the result of the method based on AdaTree showed obvious accuracy and efficient improvement compared with the method based on SVM with the kappa coefficient achieving 0.9242
78 FR 54970 - Cotton Futures Classification: Optional Classification Procedure
2013-09-09
... Service 7 CFR Part 27 [AMS-CN-13-0043] RIN 0581-AD33 Cotton Futures Classification: Optional Classification Procedure AGENCY: Agricultural Marketing Service, USDA. ACTION: Proposed rule. SUMMARY: The... optional cotton futures classification procedure--identified and known as ``registration'' by the U.S...
Classification of Indonesian quote on Twitter using Naïve Bayes
Rachmadany, A.; Pranoto, Y. M.; Gunawan; Multazam, M. T.; Nandiyanto, A. B. D.; Abdullah, A. G.; Widiaty, I.
2018-01-01
Quote is sentences made in the hope that someone can become strong personalities, individuals who always improve themselves to move forward and achieve success. Social media is a place for people to express his heart to the world that sometimes the expression of the heart is quotes. Here, the purpose of this study was to classify Indonesian quote on Twitter using Naïve Bayes. This experiment uses text classification from Twitter data written by Twitter users which are quote then classification again grouped into 6 categories (Love, Life, Motivation, Education, Religion, Others). The language used is Indonesian. The method used is Naive Bayes. The results of this experiment are a web application collection of Indonesian quote that have been classified. This classification gives the user ease in finding quote based on class or keyword. For example, when a user wants to find a 'motivation' quote, this classification would be very useful.
Improving patient safety in radiotherapy through error reporting and analysis
International Nuclear Information System (INIS)
Findlay, Ú.; Best, H.; Ottrey, M.
2016-01-01
Aim: To improve patient safety in radiotherapy (RT) through the analysis and publication of radiotherapy errors and near misses (RTE). Materials and methods: RTE are submitted on a voluntary basis by NHS RT departments throughout the UK to the National Reporting and Learning System (NRLS) or directly to Public Health England (PHE). RTE are analysed by PHE staff using frequency trend analysis based on the classification and pathway coding from Towards Safer Radiotherapy (TSRT). PHE in conjunction with the Patient Safety in Radiotherapy Steering Group publish learning from these events, on a triannual and summarised on a biennial basis, so their occurrence might be mitigated. Results: Since the introduction of this initiative in 2010, over 30,000 (RTE) reports have been submitted. The number of RTE reported in each biennial cycle has grown, ranging from 680 (2010) to 12,691 (2016) RTE. The vast majority of the RTE reported are lower level events, thus not affecting the outcome of patient care. Of the level 1 and 2 incidents reported, it is known the majority of them affected only one fraction of a course of treatment. This means that corrective action could be taken over the remaining treatment fractions so the incident did not have a significant impact on the patient or the outcome of their treatment. Analysis of the RTE reports demonstrates that generation of error is not confined to one professional group or to any particular point in the pathway. It also indicates that the pattern of errors is replicated across service providers in the UK. Conclusion: Use of the terminology, classification and coding of TSRT, together with implementation of the national voluntary reporting system described within this report, allows clinical departments to compare their local analysis to the national picture. Further opportunities to improve learning from this dataset must be exploited through development of the analysis and development of proactive risk management strategies
A Comprehensive Radial Velocity Error Budget for Next Generation Doppler Spectrometers
Halverson, Samuel; Ryan, Terrien; Mahadevan, Suvrath; Roy, Arpita; Bender, Chad; Stefansson, Guomundur Kari; Monson, Andrew; Levi, Eric; Hearty, Fred; Blake, Cullen;
2016-01-01
We describe a detailed radial velocity error budget for the NASA-NSF Extreme Precision Doppler Spectrometer instrument concept NEID (NN-explore Exoplanet Investigations with Doppler spectroscopy). Such an instrument performance budget is a necessity for both identifying the variety of noise sources currently limiting Doppler measurements, and estimating the achievable performance of next generation exoplanet hunting Doppler spectrometers. For these instruments, no single source of instrumental error is expected to set the overall measurement floor. Rather, the overall instrumental measurement precision is set by the contribution of many individual error sources. We use a combination of numerical simulations, educated estimates based on published materials, extrapolations of physical models, results from laboratory measurements of spectroscopic subsystems, and informed upper limits for a variety of error sources to identify likely sources of systematic error and construct our global instrument performance error budget. While natively focused on the performance of the NEID instrument, this modular performance budget is immediately adaptable to a number of current and future instruments. Such an approach is an important step in charting a path towards improving Doppler measurement precisions to the levels necessary for discovering Earth-like planets.
Vinay BC; Nikhitha MK; Patel Sunil B
2015-01-01
In this present review article, regarding medication errors its definition, medication error problem, types of medication errors, common causes of medication errors, monitoring medication errors, consequences of medication errors, prevention of medication error and managing medication errors have been explained neatly and legibly with proper tables which is easy to understand.
Hwang, Jae Joon; Kim, Kee-Deog; Park, Hyok; Park, Chang Seo; Jeong, Ho-Gul
2014-01-01
Superimposition has been used as a method to evaluate the changes of orthodontic or orthopedic treatment in the dental field. With the introduction of cone beam CT (CBCT), evaluating 3 dimensional changes after treatment became possible by superimposition. 4 point plane orientation is one of the simplest ways to achieve superimposition of 3 dimensional images. To find factors influencing superimposition error of cephalometric landmarks by 4 point plane orientation method and to evaluate the reproducibility of cephalometric landmarks for analyzing superimposition error, 20 patients were analyzed who had normal skeletal and occlusal relationship and took CBCT for diagnosis of temporomandibular disorder. The nasion, sella turcica, basion and midpoint between the left and the right most posterior point of the lesser wing of sphenoidal bone were used to define a three-dimensional (3D) anatomical reference co-ordinate system. Another 15 reference cephalometric points were also determined three times in the same image. Reorientation error of each landmark could be explained substantially (23%) by linear regression model, which consists of 3 factors describing position of each landmark towards reference axes and locating error. 4 point plane orientation system may produce an amount of reorientation error that may vary according to the perpendicular distance between the landmark and the x-axis; the reorientation error also increases as the locating error and shift of reference axes viewed from each landmark increases. Therefore, in order to reduce the reorientation error, accuracy of all landmarks including the reference points is important. Construction of the regression model using reference points of greater precision is required for the clinical application of this model.
Aspect-Aided Dynamic Non-Negative Sparse Representation-Based Microwave Image Classification
Directory of Open Access Journals (Sweden)
Xinzheng Zhang
2016-09-01
Full Text Available Classification of target microwave images is an important application in much areas such as security, surveillance, etc. With respect to the task of microwave image classification, a recognition algorithm based on aspect-aided dynamic non-negative least square (ADNNLS sparse representation is proposed. Firstly, an aspect sector is determined, the center of which is the estimated aspect angle of the testing sample. The training samples in the aspect sector are divided into active atoms and inactive atoms by smooth self-representative learning. Secondly, for each testing sample, the corresponding active atoms are selected dynamically, thereby establishing dynamic dictionary. Thirdly, the testing sample is represented with ℓ 1 -regularized non-negative sparse representation under the corresponding dynamic dictionary. Finally, the class label of the testing sample is identified by use of the minimum reconstruction error. Verification of the proposed algorithm was conducted using the Moving and Stationary Target Acquisition and Recognition (MSTAR database which was acquired by synthetic aperture radar. Experiment results validated that the proposed approach was able to capture the local aspect characteristics of microwave images effectively, thereby improving the classification performance.
Hoede, C.; Li, Z.
2001-01-01
In coding theory the problem of decoding focuses on error vectors. In the simplest situation code words are $(0,1)$-vectors, as are the received messages and the error vectors. Comparison of a received word with the code words yields a set of error vectors. In deciding on the original code word,
The Sample Size Influence in the Accuracy of the Image Classification of the Remote Sensing
Directory of Open Access Journals (Sweden)
Thomaz C. e C. da Costa
2004-12-01
Full Text Available Landuse/landcover maps produced by classification of remote sensing images incorporate uncertainty. This uncertainty is measured by accuracy indices using reference samples. The size of the reference sample is defined by approximation by a binomial function without the use of a pilot sample. This way the accuracy are not estimated, but fixed a priori. In case of divergency between the estimated and a priori accuracy the error of the sampling will deviate from the expected error. The size using pilot sample (theorically correct procedure justify when haven´t estimate of accuracy for work area, referent the product remote sensing utility.
Directory of Open Access Journals (Sweden)
Boulesteix Anne-Laure
2009-12-01
Full Text Available Abstract Background In biometric practice, researchers often apply a large number of different methods in a "trial-and-error" strategy to get as much as possible out of their data and, due to publication pressure or pressure from the consulting customer, present only the most favorable results. This strategy may induce a substantial optimistic bias in prediction error estimation, which is quantitatively assessed in the present manuscript. The focus of our work is on class prediction based on high-dimensional data (e.g. microarray data, since such analyses are particularly exposed to this kind of bias. Methods In our study we consider a total of 124 variants of classifiers (possibly including variable selection or tuning steps within a cross-validation evaluation scheme. The classifiers are applied to original and modified real microarray data sets, some of which are obtained by randomly permuting the class labels to mimic non-informative predictors while preserving their correlation structure. Results We assess the minimal misclassification rate over the different variants of classifiers in order to quantify the bias arising when the optimal classifier is selected a posteriori in a data-driven manner. The bias resulting from the parameter tuning (including gene selection parameters as a special case and the bias resulting from the choice of the classification method are examined both separately and jointly. Conclusions The median minimal error rate over the investigated classifiers was as low as 31% and 41% based on permuted uninformative predictors from studies on colon cancer and prostate cancer, respectively. We conclude that the strategy to present only the optimal result is not acceptable because it yields a substantial bias in error rate estimation, and suggest alternative approaches for properly reporting classification accuracy.
Fuzzy Classification of High Resolution Remote Sensing Scenes Using Visual Attention Features
Directory of Open Access Journals (Sweden)
Linyi Li
2017-01-01
Full Text Available In recent years the spatial resolutions of remote sensing images have been improved greatly. However, a higher spatial resolution image does not always lead to a better result of automatic scene classification. Visual attention is an important characteristic of the human visual system, which can effectively help to classify remote sensing scenes. In this study, a novel visual attention feature extraction algorithm was proposed, which extracted visual attention features through a multiscale process. And a fuzzy classification method using visual attention features (FC-VAF was developed to perform high resolution remote sensing scene classification. FC-VAF was evaluated by using remote sensing scenes from widely used high resolution remote sensing images, including IKONOS, QuickBird, and ZY-3 images. FC-VAF achieved more accurate classification results than the others according to the quantitative accuracy evaluation indices. We also discussed the role and impacts of different decomposition levels and different wavelets on the classification accuracy. FC-VAF improves the accuracy of high resolution scene classification and therefore advances the research of digital image analysis and the applications of high resolution remote sensing images.
[Monitoring medication errors in an internal medicine service].
Smith, Ann-Loren M; Ruiz, Inés A; Jirón, Marcela A
2014-01-01
Patients admitted to internal medicine services receive multiple drugs and thus are at risk of medication errors. To determine the frequency of medication errors (ME) among patients admitted to an internal medicine service of a high complexity hospital. A prospective observational study conducted in 225 patients admitted to an internal medicine service. Each stage of drug utilization system (prescription, transcription, dispensing, preparation and administration) was directly observed by trained pharmacists not related to hospital staff during three months. ME were described and categorized according to the National Coordinating Council for Medication Error Reporting and Prevention. In each stage of medication use, the frequency of ME and their characteristics were determined. A total of 454 drugs were prescribed to the studied patients. In 138 (30,4%) indications, at least one ME occurred, involving 67 (29,8%) patients. Twenty four percent of detected ME occurred during administration, mainly due to wrong time schedules. Anticoagulants were the therapeutic group with the highest occurrence of ME. At least one ME occurred in approximately one third of patients studied, especially during the administration stage. These errors could affect the medication safety and avoid achieving therapeutic goals. Strategies to improve the quality and safe use of medications can be implemented using this information.
The Determinants of Early Refractive Error on School-Going Chinese Children
Directory of Open Access Journals (Sweden)
K. Jayaraman
2016-04-01
Full Text Available Refractive error is a common social issue in every walks of human life, and its prevalence recorded the highest among Chinese population, particularly among people living in southern China, Hong Kong, Thailand, Singapore, and Malaysia. Refractive error is the simplest disorder to treat and supposed to cost the effective health care intervention. The present study included 168 Chinese school-going children aged 10 to 12 years; they were selected from different schools of urban Malaysia. It was surprising to see that 112 (66.7% children had the early onset of refractive error; refractive error was also detected late among the primary school or secondary school students. The findings revealed that the determinants of refractive error among Chinese children were personal achievements and machine dependence. The possible reasons for the above significant factors emerged could be attributed to the inbuilt culture and traditions of Chinese parents who insist that their children should be hardworking and focus on school subjects so that their parents allow them to use luxury electronic devices.
Classification of refrigerants; Classification des fluides frigorigenes
Energy Technology Data Exchange (ETDEWEB)
NONE
2001-07-01
This document was made from the US standard ANSI/ASHRAE 34 published in 2001 and entitled 'designation and safety classification of refrigerants'. This classification allows to clearly organize in an international way the overall refrigerants used in the world thanks to a codification of the refrigerants in correspondence with their chemical composition. This note explains this codification: prefix, suffixes (hydrocarbons and derived fluids, azeotropic and non-azeotropic mixtures, various organic compounds, non-organic compounds), safety classification (toxicity, flammability, case of mixtures). (J.S.)
Machine-learning-assisted correction of correlated qubit errors in a topological code
Directory of Open Access Journals (Sweden)
Paul Baireuther
2018-01-01
Full Text Available A fault-tolerant quantum computation requires an efficient means to detect and correct errors that accumulate in encoded quantum information. In the context of machine learning, neural networks are a promising new approach to quantum error correction. Here we show that a recurrent neural network can be trained, using only experimentally accessible data, to detect errors in a widely used topological code, the surface code, with a performance above that of the established minimum-weight perfect matching (or blossom decoder. The performance gain is achieved because the neural network decoder can detect correlations between bit-flip (X and phase-flip (Z errors. The machine learning algorithm adapts to the physical system, hence no noise model is needed. The long short-term memory layers of the recurrent neural network maintain their performance over a large number of quantum error correction cycles, making it a practical decoder for forthcoming experimental realizations of the surface code.
Chestek, Cynthia A.; Gilja, Vikash; Blabe, Christine H.; Foster, Brett L.; Shenoy, Krishna V.; Parvizi, Josef; Henderson, Jaimie M.
2013-04-01
Objective. Brain-machine interface systems translate recorded neural signals into command signals for assistive technology. In individuals with upper limb amputation or cervical spinal cord injury, the restoration of a useful hand grasp could significantly improve daily function. We sought to determine if electrocorticographic (ECoG) signals contain sufficient information to select among multiple hand postures for a prosthetic hand, orthotic, or functional electrical stimulation system.Approach. We recorded ECoG signals from subdural macro- and microelectrodes implanted in motor areas of three participants who were undergoing inpatient monitoring for diagnosis and treatment of intractable epilepsy. Participants performed five distinct isometric hand postures, as well as four distinct finger movements. Several control experiments were attempted in order to remove sensory information from the classification results. Online experiments were performed with two participants. Main results. Classification rates were 68%, 84% and 81% for correct identification of 5 isometric hand postures offline. Using 3 potential controls for removing sensory signals, error rates were approximately doubled on average (2.1×). A similar increase in errors (2.6×) was noted when the participant was asked to make simultaneous wrist movements along with the hand postures. In online experiments, fist versus rest was successfully classified on 97% of trials; the classification output drove a prosthetic hand. Online classification performance for a larger number of hand postures remained above chance, but substantially below offline performance. In addition, the long integration windows used would preclude the use of decoded signals for control of a BCI system. Significance. These results suggest that ECoG is a plausible source of command signals for prosthetic grasp selection. Overall, avenues remain for improvement through better electrode designs and placement, better participant training
Performance-scalable volumetric data classification for online industrial inspection
Abraham, Aby J.; Sadki, Mustapha; Lea, R. M.
2002-03-01
Non-intrusive inspection and non-destructive testing of manufactured objects with complex internal structures typically requires the enhancement, analysis and visualization of high-resolution volumetric data. Given the increasing availability of fast 3D scanning technology (e.g. cone-beam CT), enabling on-line detection and accurate discrimination of components or sub-structures, the inherent complexity of classification algorithms inevitably leads to throughput bottlenecks. Indeed, whereas typical inspection throughput requirements range from 1 to 1000 volumes per hour, depending on density and resolution, current computational capability is one to two orders-of-magnitude less. Accordingly, speeding up classification algorithms requires both reduction of algorithm complexity and acceleration of computer performance. A shape-based classification algorithm, offering algorithm complexity reduction, by using ellipses as generic descriptors of solids-of-revolution, and supporting performance-scalability, by exploiting the inherent parallelism of volumetric data, is presented. A two-stage variant of the classical Hough transform is used for ellipse detection and correlation of the detected ellipses facilitates position-, scale- and orientation-invariant component classification. Performance-scalability is achieved cost-effectively by accelerating a PC host with one or more COTS (Commercial-Off-The-Shelf) PCI multiprocessor cards. Experimental results are reported to demonstrate the feasibility and cost-effectiveness of the data-parallel classification algorithm for on-line industrial inspection applications.
Texture-based classification of different gastric tumors at contrast-enhanced CT
Energy Technology Data Exchange (ETDEWEB)
Ba-Ssalamah, Ahmed, E-mail: ahmed.ba-ssalamah@meduniwien.ac.at [Department of Radiology, Medical University of Vienna (Austria); Muin, Dina; Schernthaner, Ruediger; Kulinna-Cosentini, Christiana; Bastati, Nina [Department of Radiology, Medical University of Vienna (Austria); Stift, Judith [Department of Pathology, Medical University of Vienna (Austria); Gore, Richard [Department of Radiology, University of Chicago Pritzker School of Medicine, Chicago, IL (United States); Mayerhoefer, Marius E. [Department of Radiology, Medical University of Vienna (Austria)
2013-10-01
Purpose: To determine the feasibility of texture analysis for the classification of gastric adenocarcinoma, lymphoma, and gastrointestinal stromal tumors on contrast-enhanced hydrodynamic-MDCT images. Materials and methods: The arterial phase scans of 47 patients with adenocarcinoma (AC) and a histologic tumor grade of [AC-G1, n = 4, G1, n = 4; AC-G2, n = 7; AC-G3, n = 16]; GIST, n = 15; and lymphoma, n = 5, and the venous phase scans of 48 patients with AC-G1, n = 3; AC-G2, n = 6; AC-G3, n = 14; GIST, n = 17; lymphoma, n = 8, were retrospectively reviewed. Based on regions of interest, texture analysis was performed, and features derived from the gray-level histogram, run-length and co-occurrence matrix, absolute gradient, autoregressive model, and wavelet transform were calculated. Fisher coefficients, probability of classification error, average correlation coefficients, and mutual information coefficients were used to create combinations of texture features that were optimized for tumor differentiation. Linear discriminant analysis in combination with a k-nearest neighbor classifier was used for tumor classification. Results: On arterial-phase scans, texture-based lesion classification was highly successful in differentiating between AC and lymphoma, and GIST and lymphoma, with misclassification rates of 3.1% and 0%, respectively. On venous-phase scans, texture-based classification was slightly less successful for AC vs. lymphoma (9.7% misclassification) and GIST vs. lymphoma (8% misclassification), but enabled the differentiation between AC and GIST (10% misclassification), and between the different grades of AC (4.4% misclassification). No texture feature combination was able to adequately distinguish between all three tumor types. Conclusion: Classification of different gastric tumors based on textural information may aid radiologists in establishing the correct diagnosis, at least in cases where the differential diagnosis can be narrowed down to two
Texture-based classification of different gastric tumors at contrast-enhanced CT
International Nuclear Information System (INIS)
Ba-Ssalamah, Ahmed; Muin, Dina; Schernthaner, Ruediger; Kulinna-Cosentini, Christiana; Bastati, Nina; Stift, Judith; Gore, Richard; Mayerhoefer, Marius E.
2013-01-01
Purpose: To determine the feasibility of texture analysis for the classification of gastric adenocarcinoma, lymphoma, and gastrointestinal stromal tumors on contrast-enhanced hydrodynamic-MDCT images. Materials and methods: The arterial phase scans of 47 patients with adenocarcinoma (AC) and a histologic tumor grade of [AC-G1, n = 4, G1, n = 4; AC-G2, n = 7; AC-G3, n = 16]; GIST, n = 15; and lymphoma, n = 5, and the venous phase scans of 48 patients with AC-G1, n = 3; AC-G2, n = 6; AC-G3, n = 14; GIST, n = 17; lymphoma, n = 8, were retrospectively reviewed. Based on regions of interest, texture analysis was performed, and features derived from the gray-level histogram, run-length and co-occurrence matrix, absolute gradient, autoregressive model, and wavelet transform were calculated. Fisher coefficients, probability of classification error, average correlation coefficients, and mutual information coefficients were used to create combinations of texture features that were optimized for tumor differentiation. Linear discriminant analysis in combination with a k-nearest neighbor classifier was used for tumor classification. Results: On arterial-phase scans, texture-based lesion classification was highly successful in differentiating between AC and lymphoma, and GIST and lymphoma, with misclassification rates of 3.1% and 0%, respectively. On venous-phase scans, texture-based classification was slightly less successful for AC vs. lymphoma (9.7% misclassification) and GIST vs. lymphoma (8% misclassification), but enabled the differentiation between AC and GIST (10% misclassification), and between the different grades of AC (4.4% misclassification). No texture feature combination was able to adequately distinguish between all three tumor types. Conclusion: Classification of different gastric tumors based on textural information may aid radiologists in establishing the correct diagnosis, at least in cases where the differential diagnosis can be narrowed down to two
Classification of resistance to passive motion using minimum probability of error criterion.
Chan, H C; Manry, M T; Kondraske, G V
1987-01-01
Neurologists diagnose many muscular and nerve disorders by classifying the resistance to passive motion of patients' limbs. Over the past several years, a computer-based instrument has been developed for automated measurement and parameterization of this resistance. In the device, a voluntarily relaxed lower extremity is moved at constant velocity by a motorized driver. The torque exerted on the extremity by the machine is sampled, along with the angle of the extremity. In this paper a computerized technique is described for classifying a patient's condition as 'Normal' or 'Parkinson disease' (rigidity), from the torque versus angle curve for the knee joint. A Legendre polynomial, fit to the curve, is used to calculate a set of eight normally distributed features of the curve. The minimum probability of error approach is used to classify the curve as being from a normal or Parkinson disease patient. Data collected from 44 different subjects was processes and the results were compared with an independent physician's subjective assessment of rigidity. There is agreement in better than 95% of the cases, when all of the features are used.
An Extended Spectral-Spatial Classification Approach for Hyperspectral Data
Akbari, D.
2017-11-01
In this paper an extended classification approach for hyperspectral imagery based on both spectral and spatial information is proposed. The spatial information is obtained by an enhanced marker-based minimum spanning forest (MSF) algorithm. Three different methods of dimension reduction are first used to obtain the subspace of hyperspectral data: (1) unsupervised feature extraction methods including principal component analysis (PCA), independent component analysis (ICA), and minimum noise fraction (MNF); (2) supervised feature extraction including decision boundary feature extraction (DBFE), discriminate analysis feature extraction (DAFE), and nonparametric weighted feature extraction (NWFE); (3) genetic algorithm (GA). The spectral features obtained are then fed into the enhanced marker-based MSF classification algorithm. In the enhanced MSF algorithm, the markers are extracted from the classification maps obtained by both SVM and watershed segmentation algorithm. To evaluate the proposed approach, the Pavia University hyperspectral data is tested. Experimental results show that the proposed approach using GA achieves an approximately 8 % overall accuracy higher than the original MSF-based algorithm.
International Nuclear Information System (INIS)
Jerebko, Anna K.; Summers, Ronald M.; Malley, James D.; Franaszek, Marek; Johnson, C. Daniel
2003-01-01
Detection of colonic polyps in CT colonography is problematic due to complexities of polyp shape and the surface of the normal colon. Published results indicate the feasibility of computer-aided detection of polyps but better classifiers are needed to improve specificity. In this paper we compare the classification results of two approaches: neural networks and recursive binary trees. As our starting point we collect surface geometry information from three-dimensional reconstruction of the colon, followed by a filter based on selected variables such as region density, Gaussian and average curvature and sphericity. The filter returns sites that are candidate polyps, based on earlier work using detection thresholds, to which the neural nets or the binary trees are applied. A data set of 39 polyps from 3 to 25 mm in size was used in our investigation. For both neural net and binary trees we use tenfold cross-validation to better estimate the true error rates. The backpropagation neural net with one hidden layer trained with Levenberg-Marquardt algorithm achieved the best results: sensitivity 90% and specificity 95% with 16 false positives per study
CALIBRATION ERRORS IN THE CAVITY BEAM POSITION MONITOR SYSTEM AT THE ATF2
Cullinan, F; Joshi, N; Lyapin, A
2011-01-01
It has been shown at the Accelerator Test Facility at KEK, that it is possible to run a system of 37 cavity beam position monitors (BPMs) and achieve high working resolution. However, stability of the calibration constants (position scale and radio frequency (RF) phase) over a three/four week running period is yet to be demonstrated. During the calibration procedure, random beam jitter gives rise to a statistical error in the position scale and slow orbit drift in position and tilt causes systematic errors in both the position scale and RF phase. These errors are dominant and have been evaluated for each BPM. The results are compared with the errors expected after a tested method of beam jitter subtraction has been applied.
Seizure classification in EEG signals utilizing Hilbert-Huang transform
Directory of Open Access Journals (Sweden)
Abdulhay Enas W
2011-05-01
Full Text Available Abstract Background Classification method capable of recognizing abnormal activities of the brain functionality are either brain imaging or brain signal analysis. The abnormal activity of interest in this study is characterized by a disturbance caused by changes in neuronal electrochemical activity that results in abnormal synchronous discharges. The method aims at helping physicians discriminate between healthy and seizure electroencephalographic (EEG signals. Method Discrimination in this work is achieved by analyzing EEG signals obtained from freely accessible databases. MATLAB has been used to implement and test the proposed classification algorithm. The analysis in question presents a classification of normal and ictal activities using a feature relied on Hilbert-Huang Transform. Through this method, information related to the intrinsic functions contained in the EEG signal has been extracted to track the local amplitude and the frequency of the signal. Based on this local information, weighted frequencies are calculated and a comparison between ictal and seizure-free determinant intrinsic functions is then performed. Methods of comparison used are the t-test and the Euclidean clustering. Results The t-test results in a P-value Conclusion An original tool for EEG signal processing giving physicians the possibility to diagnose brain functionality abnormalities is presented in this paper. The proposed system bears the potential of providing several credible benefits such as fast diagnosis, high accuracy, good sensitivity and specificity, time saving and user friendly. Furthermore, the classification of mode mixing can be achieved using the extracted instantaneous information of every IMF, but it would be most likely a hard task if only the average value is used. Extra benefits of this proposed system include low cost, and ease of interface. All of that indicate the usefulness of the tool and its use as an efficient diagnostic tool.
Seizure classification in EEG signals utilizing Hilbert-Huang transform.
Oweis, Rami J; Abdulhay, Enas W
2011-05-24
Classification method capable of recognizing abnormal activities of the brain functionality are either brain imaging or brain signal analysis. The abnormal activity of interest in this study is characterized by a disturbance caused by changes in neuronal electrochemical activity that results in abnormal synchronous discharges. The method aims at helping physicians discriminate between healthy and seizure electroencephalographic (EEG) signals. Discrimination in this work is achieved by analyzing EEG signals obtained from freely accessible databases. MATLAB has been used to implement and test the proposed classification algorithm. The analysis in question presents a classification of normal and ictal activities using a feature relied on Hilbert-Huang Transform. Through this method, information related to the intrinsic functions contained in the EEG signal has been extracted to track the local amplitude and the frequency of the signal. Based on this local information, weighted frequencies are calculated and a comparison between ictal and seizure-free determinant intrinsic functions is then performed. Methods of comparison used are the t-test and the Euclidean clustering. The t-test results in a P-value with respect to its fast response and ease to use. An original tool for EEG signal processing giving physicians the possibility to diagnose brain functionality abnormalities is presented in this paper. The proposed system bears the potential of providing several credible benefits such as fast diagnosis, high accuracy, good sensitivity and specificity, time saving and user friendly. Furthermore, the classification of mode mixing can be achieved using the extracted instantaneous information of every IMF, but it would be most likely a hard task if only the average value is used. Extra benefits of this proposed system include low cost, and ease of interface. All of that indicate the usefulness of the tool and its use as an efficient diagnostic tool.
Emotion of Physiological Signals Classification Based on TS Feature Selection
Institute of Scientific and Technical Information of China (English)
Wang Yujing; Mo Jianlin
2015-01-01
This paper propose a method of TS-MLP about emotion recognition of physiological signal.It can recognize emotion successfully by Tabu search which selects features of emotion’s physiological signals and multilayer perceptron that is used to classify emotion.Simulation shows that it has achieved good emotion classification performance.
Medication errors with electronic prescribing (eP): Two views of the same picture
2010-01-01
Background Quantitative prospective methods are widely used to evaluate the impact of new technologies such as electronic prescribing (eP) on medication errors. However, they are labour-intensive and it is not always feasible to obtain pre-intervention data. Our objective was to compare the eP medication error picture obtained with retrospective quantitative and qualitative methods. Methods The study was carried out at one English district general hospital approximately two years after implementation of an integrated electronic prescribing, administration and records system. Quantitative: A structured retrospective analysis was carried out of clinical records and medication orders for 75 randomly selected patients admitted to three wards (medicine, surgery and paediatrics) six months after eP implementation. Qualitative: Eight doctors, 6 nurses, 8 pharmacy staff and 4 other staff at senior, middle and junior grades, and 19 adult patients on acute surgical and medical wards were interviewed. Staff interviews explored experiences of developing and working with the system; patient interviews focused on experiences of medicine prescribing and administration on the ward. Interview transcripts were searched systematically for accounts of medication incidents. A classification scheme was developed and applied to the errors identified in the records review. Results The two approaches produced similar pictures of the drug use process. Interviews identified types of error identified in the retrospective notes review plus two eP-specific errors which were not detected by record review. Interview data took less time to collect than record review, and provided rich data on the prescribing process, and reasons for delays or non-administration of medicines, including "once only" orders and "as required" medicines. Conclusions The qualitative approach provided more understanding of processes, and some insights into why medication errors can happen. The method is cost-effective and
Faust, Kevin; Xie, Quin; Han, Dominick; Goyle, Kartikay; Volynskaya, Zoya; Djuric, Ugljesa; Diamandis, Phedias
2018-05-16
There is growing interest in utilizing artificial intelligence, and particularly deep learning, for computer vision in histopathology. While accumulating studies highlight expert-level performance of convolutional neural networks (CNNs) on focused classification tasks, most studies rely on probability distribution scores with empirically defined cutoff values based on post-hoc analysis. More generalizable tools that allow humans to visualize histology-based deep learning inferences and decision making are scarce. Here, we leverage t-distributed Stochastic Neighbor Embedding (t-SNE) to reduce dimensionality and depict how CNNs organize histomorphologic information. Unique to our workflow, we develop a quantitative and transparent approach to visualizing classification decisions prior to softmax compression. By discretizing the relationships between classes on the t-SNE plot, we show we can super-impose randomly sampled regions of test images and use their distribution to render statistically-driven classifications. Therefore, in addition to providing intuitive outputs for human review, this visual approach can carry out automated and objective multi-class classifications similar to more traditional and less-transparent categorical probability distribution scores. Importantly, this novel classification approach is driven by a priori statistically defined cutoffs. It therefore serves as a generalizable classification and anomaly detection tool less reliant on post-hoc tuning. Routine incorporation of this convenient approach for quantitative visualization and error reduction in histopathology aims to accelerate early adoption of CNNs into generalized real-world applications where unanticipated and previously untrained classes are often encountered.
Intraoperative visualization and assessment of electromagnetic tracking error
Harish, Vinyas; Ungi, Tamas; Lasso, Andras; MacDonald, Andrew; Nanji, Sulaiman; Fichtinger, Gabor
2015-03-01
Electromagnetic tracking allows for increased flexibility in designing image-guided interventions, however it is well understood that electromagnetic tracking is prone to error. Visualization and assessment of the tracking error should take place in the operating room with minimal interference with the clinical procedure. The goal was to achieve this ideal in an open-source software implementation in a plug and play manner, without requiring programming from the user. We use optical tracking as a ground truth. An electromagnetic sensor and optical markers are mounted onto a stylus device, pivot calibrated for both trackers. Electromagnetic tracking error is defined as difference of tool tip position between electromagnetic and optical readings. Multiple measurements are interpolated into the thin-plate B-spline transform visualized in real time using 3D Slicer. All tracked devices are used in a plug and play manner through the open-source SlicerIGT and PLUS extensions of the 3D Slicer platform. Tracking error was measured multiple times to assess reproducibility of the method, both with and without placing ferromagnetic objects in the workspace. Results from exhaustive grid sampling and freehand sampling were similar, indicating that a quick freehand sampling is sufficient to detect unexpected or excessive field distortion in the operating room. The software is available as a plug-in for the 3D Slicer platforms. Results demonstrate potential for visualizing electromagnetic tracking error in real time for intraoperative environments in feasibility clinical trials in image-guided interventions.
The error in total error reduction.
Witnauer, James E; Urcelay, Gonzalo P; Miller, Ralph R
2014-02-01
Most models of human and animal learning assume that learning is proportional to the discrepancy between a delivered outcome and the outcome predicted by all cues present during that trial (i.e., total error across a stimulus compound). This total error reduction (TER) view has been implemented in connectionist and artificial neural network models to describe the conditions under which weights between units change. Electrophysiological work has revealed that the activity of dopamine neurons is correlated with the total error signal in models of reward learning. Similar neural mechanisms presumably support fear conditioning, human contingency learning, and other types of learning. Using a computational modeling approach, we compared several TER models of associative learning to an alternative model that rejects the TER assumption in favor of local error reduction (LER), which assumes that learning about each cue is proportional to the discrepancy between the delivered outcome and the outcome predicted by that specific cue on that trial. The LER model provided a better fit to the reviewed data than the TER models. Given the superiority of the LER model with the present data sets, acceptance of TER should be tempered. Copyright © 2013 Elsevier Inc. All rights reserved.
adabag: An R Package for Classification with Boosting and Bagging
Directory of Open Access Journals (Sweden)
Esteban Alfaro
2013-09-01
Full Text Available Boosting and bagging are two widely used ensemble methods for classification. Their common goal is to improve the accuracy of a classifier combining single classifiers which are slightly better than random guessing. Among the family of boosting algorithms, AdaBoost (adaptive boosting is the best known, although it is suitable only for dichotomous tasks. AdaBoost.M1 and SAMME (stagewise additive modeling using a multi-class exponential loss function are two easy and natural extensions to the general case of two or more classes. In this paper, the adabag R package is introduced. This version implements AdaBoost.M1, SAMME and bagging algorithms with classification trees as base classifiers. Once the ensembles have been trained, they can be used to predict the class of new samples. The accuracy of these classifiers can be estimated in a separated data set or through cross validation. Moreover, the evolution of the error as the ensemble grows can be analysed and the ensemble can be pruned. In addition, the margin in the class prediction and the probability of each class for the observations can be calculated. Finally, several classic examples in classification literature are shown to illustrate the use of this package.
Flow regime classification in air-magnetic fluid two-phase flow.
Kuwahara, T; De Vuyst, F; Yamaguchi, H
2008-05-21
A new experimental/numerical technique of classification of flow regimes (flow patterns) in air-magnetic fluid two-phase flow is proposed in the present paper. The proposed technique utilizes the electromagnetic induction to obtain time-series signals of the electromotive force, allowing us to make a non-contact measurement. Firstly, an experiment is carried out to obtain the time-series signals in a vertical upward air-magnetic fluid two-phase flow. The signals obtained are first treated using two kinds of wavelet transforms. The data sets treated are then used as input vectors for an artificial neural network (ANN) with supervised training. In the present study, flow regimes are classified into bubbly, slug, churn and annular flows, which are generally the main flow regimes. To validate the flow regimes, a visualization experiment is also performed with a glycerin solution that has roughly the same physical properties, i.e., kinetic viscosity and surface tension, as a magnetic fluid used in the present study. The flow regimes from the visualization are used as targets in an ANN and also used in the estimation of the accuracy of the present method. As a result, ANNs using radial basis functions are shown to be the most appropriate for the present classification of flow regimes, leading to small classification errors.
Directory of Open Access Journals (Sweden)
C. Zhang
2017-09-01
Full Text Available Recent advances in remote sensing have witnessed a great amount of very high resolution (VHR images acquired at sub-metre spatial resolution. These VHR remotely sensed data has post enormous challenges in processing, analysing and classifying them effectively due to the high spatial complexity and heterogeneity. Although many computer-aid classification methods that based on machine learning approaches have been developed over the past decades, most of them are developed toward pixel level spectral differentiation, e.g. Multi-Layer Perceptron (MLP, which are unable to exploit abundant spatial details within VHR images. This paper introduced a rough set model as a general framework to objectively characterize the uncertainty in CNN classification results, and further partition them into correctness and incorrectness on the map. The correct classification regions of CNN were trusted and maintained, whereas the misclassification areas were reclassified using a decision tree with both CNN and MLP. The effectiveness of the proposed rough set decision tree based MLP-CNN was tested using an urban area at Bournemouth, United Kingdom. The MLP-CNN, well capturing the complementarity between CNN and MLP through the rough set based decision tree, achieved the best classification performance both visually and numerically. Therefore, this research paves the way to achieve fully automatic and effective VHR image classification.
Characteristics of pediatric chemotherapy medication errors in a national error reporting database.
Rinke, Michael L; Shore, Andrew D; Morlock, Laura; Hicks, Rodney W; Miller, Marlene R
2007-07-01
Little is known regarding chemotherapy medication errors in pediatrics despite studies suggesting high rates of overall pediatric medication errors. In this study, the authors examined patterns in pediatric chemotherapy errors. The authors queried the United States Pharmacopeia MEDMARX database, a national, voluntary, Internet-accessible error reporting system, for all error reports from 1999 through 2004 that involved chemotherapy medications and patients aged error reports, 85% reached the patient, and 15.6% required additional patient monitoring or therapeutic intervention. Forty-eight percent of errors originated in the administering phase of medication delivery, and 30% originated in the drug-dispensing phase. Of the 387 medications cited, 39.5% were antimetabolites, 14.0% were alkylating agents, 9.3% were anthracyclines, and 9.3% were topoisomerase inhibitors. The most commonly involved chemotherapeutic agents were methotrexate (15.3%), cytarabine (12.1%), and etoposide (8.3%). The most common error types were improper dose/quantity (22.9% of 327 cited error types), wrong time (22.6%), omission error (14.1%), and wrong administration technique/wrong route (12.2%). The most common error causes were performance deficit (41.3% of 547 cited error causes), equipment and medication delivery devices (12.4%), communication (8.8%), knowledge deficit (6.8%), and written order errors (5.5%). Four of the 5 most serious errors occurred at community hospitals. Pediatric chemotherapy errors often reached the patient, potentially were harmful, and differed in quality between outpatient and inpatient areas. This study indicated which chemotherapeutic agents most often were involved in errors and that administering errors were common. Investigation is needed regarding targeted medication administration safeguards for these high-risk medications. Copyright (c) 2007 American Cancer Society.
EEG-based decoding of error-related brain activity in a real-world driving task
Zhang, H.; Chavarriaga, R.; Khaliliardali, Z.; Gheorghe, L.; Iturrate, I.; Millán, J. d. R.
2015-12-01
Objectives. Recent studies have started to explore the implementation of brain-computer interfaces (BCI) as part of driving assistant systems. The current study presents an EEG-based BCI that decodes error-related brain activity. Such information can be used, e.g., to predict driver’s intended turning direction before reaching road intersections. Approach. We executed experiments in a car simulator (N = 22) and a real car (N = 8). While subject was driving, a directional cue was shown before reaching an intersection, and we classified the presence or not of an error-related potentials from EEG to infer whether the cued direction coincided with the subject’s intention. In this protocol, the directional cue can correspond to an estimation of the driving direction provided by a driving assistance system. We analyzed ERPs elicited during normal driving and evaluated the classification performance in both offline and online tests. Results. An average classification accuracy of 0.698 ± 0.065 was obtained in offline experiments in the car simulator, while tests in the real car yielded a performance of 0.682 ± 0.059. The results were significantly higher than chance level for all cases. Online experiments led to equivalent performances in both simulated and real car driving experiments. These results support the feasibility of decoding these signals to help estimating whether the driver’s intention coincides with the advice provided by the driving assistant in a real car. Significance. The study demonstrates a BCI system in real-world driving, extending the work from previous simulated studies. As far as we know, this is the first online study in real car decoding driver’s error-related brain activity. Given the encouraging results, the paradigm could be further improved by using more sophisticated machine learning approaches and possibly be combined with applications in intelligent vehicles.
Wireless Magnetic Sensor Network for Road Traffic Monitoring and Vehicle Classification
Directory of Open Access Journals (Sweden)
Velisavljevic Vladan
2016-12-01
Full Text Available Efficiency of transportation of people and goods is playing a vital role in economic growth. A key component for enabling effective planning of transportation networks is the deployment and operation of autonomous monitoring and traffic analysis tools. For that reason, such systems have been developed to register and classify road traffic usage. In this paper, we propose a novel system for road traffic monitoring and classification based on highly energy efficient wireless magnetic sensor networks. We develop novel algorithms for vehicle speed and length estimation and vehicle classification that use multiple magnetic sensors. We also demonstrate that, using such a low-cost system with simplified installation and maintenance compared to current solutions, it is possible to achieve highly accurate estimation and a high rate of positive vehicle classification.
Directory of Open Access Journals (Sweden)
A. Ozdarici Ok
2012-09-01
training pixels are automatically selected and labelled. Thereafter, those training pixels are utilized in a traditional Maximum Likelihood Classification to classify five crop types namely; corn, tomato/pepper, rice, sugar beet, and wheat. The accuracy of the classification is evaluated in pixel-based manner with the help of a reference map including crop information of the area. Promising results are achieved for pixel-based approach. Based on the error matrices used in the evaluation, overall accuracy of the pixel-based analysis is computed as 89.31%. Similar to the overall accuracies, high individual class accuracies are obtained as well. The results point out that automatically collecting the training samples by extracting representative homogenous areas significantly increases the speed of the classification and minimizes the human interaction. The results also confirm that the proposed approach is highly appropriate for the extraction of representative homogenous training areas.
International Nuclear Information System (INIS)
Benkirane, A.; Auger, G.; Chbihi, A.; Bloyet, D.; Plagnol, E.
1994-01-01
This paper presents an original approach to solve an automatic data classification problem by means of image processing techniques. The classification is achieved using image segmentation techniques for extracting the meaningful classes. Two types of information are merged for this purpose: the information contained in experimental images and a priori information derived from underlying physics (and adapted to image segmentation problem). This data fusion is widely used at different stages of the segmentation process. This approach yields interesting results in terms of segmentation performances, even in very noisy cases. Satisfactory classification results are obtained in cases where more ''classical'' automatic data classification methods fail. (authors). 25 refs., 14 figs., 1 append
Energy Technology Data Exchange (ETDEWEB)
Benkirane, A; Auger, G; Chbihi, A [Grand Accelerateur National d` Ions Lourds (GANIL), 14 - Caen (France); Bloyet, D [Caen Univ., 14 (France); Plagnol, E [Paris-11 Univ., 91 - Orsay (France). Inst. de Physique Nucleaire
1994-12-31
This paper presents an original approach to solve an automatic data classification problem by means of image processing techniques. The classification is achieved using image segmentation techniques for extracting the meaningful classes. Two types of information are merged for this purpose: the information contained in experimental images and a priori information derived from underlying physics (and adapted to image segmentation problem). This data fusion is widely used at different stages of the segmentation process. This approach yields interesting results in terms of segmentation performances, even in very noisy cases. Satisfactory classification results are obtained in cases where more ``classical`` automatic data classification methods fail. (authors). 25 refs., 14 figs., 1 append.
Mandava, Pitchaiah; Krumpelman, Chase S; Shah, Jharna N; White, Donna L; Kent, Thomas A
2013-01-01
Clinical trial outcomes often involve an ordinal scale of subjective functional assessments but the optimal way to quantify results is not clear. In stroke, the most commonly used scale, the modified Rankin Score (mRS), a range of scores ("Shift") is proposed as superior to dichotomization because of greater information transfer. The influence of known uncertainties in mRS assessment has not been quantified. We hypothesized that errors caused by uncertainties could be quantified by applying information theory. Using Shannon's model, we quantified errors of the "Shift" compared to dichotomized outcomes using published distributions of mRS uncertainties and applied this model to clinical trials. We identified 35 randomized stroke trials that met inclusion criteria. Each trial's mRS distribution was multiplied with the noise distribution from published mRS inter-rater variability to generate an error percentage for "shift" and dichotomized cut-points. For the SAINT I neuroprotectant trial, considered positive by "shift" mRS while the larger follow-up SAINT II trial was negative, we recalculated sample size required if classification uncertainty was taken into account. Considering the full mRS range, error rate was 26.1%±5.31 (Mean±SD). Error rates were lower for all dichotomizations tested using cut-points (e.g. mRS 1; 6.8%±2.89; overall pdecrease in reliability. The resultant errors need to be considered since sample size may otherwise be underestimated. In principle, we have outlined an approach to error estimation for any condition in which there are uncertainties in outcome assessment. We provide the user with programs to calculate and incorporate errors into sample size estimation.