WorldWideScience

Sample records for learning testing predictions

  1. Test Framing Generates a Stability Bias for Predictions of Learning by Causing People to Discount their Learning Beliefs

    Science.gov (United States)

    Ariel, Robert; Hines, Jarrod C.; Hertzog, Christopher

    2014-01-01

    People estimate minimal changes in learning when making predictions of learning (POLs) for future study opportunities despite later showing increased performance and an awareness of that increase (Kornell & Bjork, 2009). This phenomenon is conceptualized as a stability bias in judgments about learning. We investigated the malleability of this effect, and whether it reflected people’s underlying beliefs about learning. We manipulated prediction framing to emphasize the role of testing vs. studying on memory and directly measured beliefs about multi-trial study effects on learning by having participants construct predicted learning curves before and after the experiment. Mean POLs were more sensitive to the number of study-test opportunities when performance was framed in terms of study benefits rather than testing benefits and POLs reflected pre-existing beliefs about learning. The stability bias is partially due to framing and reflects discounted beliefs about learning benefits rather than inherent belief in the stability of performance. PMID:25067885

  2. Feedback-related brain activity predicts learning from feedback in multiple-choice testing.

    Science.gov (United States)

    Ernst, Benjamin; Steinhauser, Marco

    2012-06-01

    Different event-related potentials (ERPs) have been shown to correlate with learning from feedback in decision-making tasks and with learning in explicit memory tasks. In the present study, we investigated which ERPs predict learning from corrective feedback in a multiple-choice test, which combines elements from both paradigms. Participants worked through sets of multiple-choice items of a Swahili-German vocabulary task. Whereas the initial presentation of an item required the participants to guess the answer, corrective feedback could be used to learn the correct response. Initial analyses revealed that corrective feedback elicited components related to reinforcement learning (FRN), as well as to explicit memory processing (P300) and attention (early frontal positivity). However, only the P300 and early frontal positivity were positively correlated with successful learning from corrective feedback, whereas the FRN was even larger when learning failed. These results suggest that learning from corrective feedback crucially relies on explicit memory processing and attentional orienting to corrective feedback, rather than on reinforcement learning.

  3. Rey's Auditory Verbal Learning Test scores can be predicted from whole brain MRI in Alzheimer's disease

    Directory of Open Access Journals (Sweden)

    Elaheh Moradi

    2017-01-01

    Full Text Available Rey's Auditory Verbal Learning Test (RAVLT is a powerful neuropsychological tool for testing episodic memory, which is widely used for the cognitive assessment in dementia and pre-dementia conditions. Several studies have shown that an impairment in RAVLT scores reflect well the underlying pathology caused by Alzheimer's disease (AD, thus making RAVLT an effective early marker to detect AD in persons with memory complaints. We investigated the association between RAVLT scores (RAVLT Immediate and RAVLT Percent Forgetting and the structural brain atrophy caused by AD. The aim was to comprehensively study to what extent the RAVLT scores are predictable based on structural magnetic resonance imaging (MRI data using machine learning approaches as well as to find the most important brain regions for the estimation of RAVLT scores. For this, we built a predictive model to estimate RAVLT scores from gray matter density via elastic net penalized linear regression model. The proposed approach provided highly significant cross-validated correlation between the estimated and observed RAVLT Immediate (R = 0.50 and RAVLT Percent Forgetting (R = 0.43 in a dataset consisting of 806 AD, mild cognitive impairment (MCI or healthy subjects. In addition, the selected machine learning method provided more accurate estimates of RAVLT scores than the relevance vector regression used earlier for the estimation of RAVLT based on MRI data. The top predictors were medial temporal lobe structures and amygdala for the estimation of RAVLT Immediate and angular gyrus, hippocampus and amygdala for the estimation of RAVLT Percent Forgetting. Further, the conversion of MCI subjects to AD in 3-years could be predicted based on either observed or estimated RAVLT scores with an accuracy comparable to MRI-based biomarkers.

  4. Predictive Testing

    Science.gov (United States)

    ... your family Plan for the future Insurance and financial planning Transition for children Emergency preparedness Testing & Services Testing ... Support Genetic Disease Information Find a Support Group Financial Planning Who Should I Tell? Genetic Testing & Counseling Compensation ...

  5. Using a Comprehensive Model to Test and Predict the Factors of Online Learning Effectiveness

    Science.gov (United States)

    He, Minyan

    2013-01-01

    As online learning is an important part of higher education, the effectiveness of online learning has been tested with different methods. Although the literature regarding online learning effectiveness has been related to various factors, a more comprehensive review of the factors may result in broader understanding of online learning…

  6. Hippocampal dose volume histogram predicts Hopkins Verbal Learning Test scores after brain irradiation

    Directory of Open Access Journals (Sweden)

    Catherine Okoukoni, PhD

    2017-10-01

    Full Text Available Purpose: Radiation-induced cognitive decline is relatively common after treatment for primary and metastatic brain tumors; however, identifying dosimetric parameters that are predictive of radiation-induced cognitive decline is difficult due to the heterogeneity of patient characteristics. The memory function is especially susceptible to radiation effects after treatment. The objective of this study is to correlate volumetric radiation doses received by critical neuroanatomic structures to post–radiation therapy (RT memory impairment. Methods and materials: Between 2008 and 2011, 53 patients with primary brain malignancies were treated with conventionally fractionated RT in prospectively accrued clinical trials performed at our institution. Dose-volume histogram analysis was performed for the hippocampus, parahippocampus, amygdala, and fusiform gyrus. Hopkins Verbal Learning Test-Revised scores were obtained at least 6 months after RT. Impairment was defined as an immediate recall score ≤15. For each anatomic region, serial regression was performed to correlate volume receiving a given dose (VD(Gy with memory impairment. Results: Hippocampal V53.4Gy to V60.9Gy significantly predicted post-RT memory impairment (P < .05. Within this range, the hippocampal V55Gy was the most significant predictor (P = .004. Hippocampal V55Gy of 0%, 25%, and 50% was associated with tumor-induced impairment rates of 14.9% (95% confidence interval [CI], 7.2%-28.7%, 45.9% (95% CI, 24.7%-68.6%, and 80.6% (95% CI, 39.2%-96.4%, respectively. Conclusions: The hippocampal V55Gy is a significant predictor for impairment, and a limiting dose below 55 Gy may minimize radiation-induced cognitive impairment.

  7. Improved method for SNR prediction in machine-learning-based test

    NARCIS (Netherlands)

    Sheng, Xiaoqin; Kerkhoff, Hans G.

    2010-01-01

    This paper applies an improved method for testing the signal-to-noise ratio (SNR) of Analogue-to-Digital Converters (ADC). In previous work, a noisy and nonlinear pulse signal is exploited as the input stimulus to obtain the signature results of ADC. By applying a machine-learning-based approach,

  8. A novel test for evaluating horses' spontaneous visual attention is predictive of attention in operant learning tasks.

    Science.gov (United States)

    Rochais, C; Sébilleau, M; Houdebine, M; Bec, P; Hausberger, M; Henry, S

    2017-08-01

    Attention is described as the ability to process selectively one aspect of the environment over others. In this study, we characterized horses' spontaneous attention by designing a novel visual attention test (VAT) that is easy to apply in the animal's home environment. The test was repeated over three consecutive days and repeated again 6 months later in order to assess inter-individual variations and intra-individual stability. Different patterns of attention have been revealed: 'overall' attention when the horse merely gazed at the stimulus and 'fixed' attention characterized by fixity and orientation of at least the visual and auditory organs towards the stimulus. The individual attention characteristics remained consistent over time (after 6 months, Spearman correlation test, P attentional skills was assessed by comparing the results, for the same horses, with those obtained in both a 'classical' experimental attention test the 'five-choice serial reaction time task' (5-CSRTT) and a work situation (lunge working context). Our results revealed that (i) individual variations remained consistent across tests and (ii) the VAT attention measures were not only predictive of attentional skills but also of learning abilities. Differences appeared however between the first day of testing and the following test days: attention structure on the second day was predictive of learning abilities, attention performances in the 5-CSRRT and at work. The VAT appears as a promising easy-to-use tool to assess animals' attention characteristics and the impact of different factors of variation on attention.

  9. A novel test for evaluating horses' spontaneous visual attention is predictive of attention in operant learning tasks

    Science.gov (United States)

    Rochais, C.; Sébilleau, M.; Houdebine, M.; Bec, P.; Hausberger, M.; Henry, S.

    2017-08-01

    Attention is described as the ability to process selectively one aspect of the environment over others. In this study, we characterized horses' spontaneous attention by designing a novel visual attention test (VAT) that is easy to apply in the animal's home environment. The test was repeated over three consecutive days and repeated again 6 months later in order to assess inter-individual variations and intra-individual stability. Different patterns of attention have been revealed: `overall' attention when the horse merely gazed at the stimulus and `fixed' attention characterized by fixity and orientation of at least the visual and auditory organs towards the stimulus. The individual attention characteristics remained consistent over time (after 6 months, Spearman correlation test, P work situation (lunge working context). Our results revealed that (i) individual variations remained consistent across tests and (ii) the VAT attention measures were not only predictive of attentional skills but also of learning abilities. Differences appeared however between the first day of testing and the following test days: attention structure on the second day was predictive of learning abilities, attention performances in the 5-CSRRT and at work. The VAT appears as a promising easy-to-use tool to assess animals' attention characteristics and the impact of different factors of variation on attention.

  10. Predicting diabetes mellitus using SMOTE and ensemble machine learning approach: The Henry Ford ExercIse Testing (FIT) project.

    Science.gov (United States)

    Alghamdi, Manal; Al-Mallah, Mouaz; Keteyian, Steven; Brawner, Clinton; Ehrman, Jonathan; Sakr, Sherif

    2017-01-01

    Machine learning is becoming a popular and important approach in the field of medical research. In this study, we investigate the relative performance of various machine learning methods such as Decision Tree, Naïve Bayes, Logistic Regression, Logistic Model Tree and Random Forests for predicting incident diabetes using medical records of cardiorespiratory fitness. In addition, we apply different techniques to uncover potential predictors of diabetes. This FIT project study used data of 32,555 patients who are free of any known coronary artery disease or heart failure who underwent clinician-referred exercise treadmill stress testing at Henry Ford Health Systems between 1991 and 2009 and had a complete 5-year follow-up. At the completion of the fifth year, 5,099 of those patients have developed diabetes. The dataset contained 62 attributes classified into four categories: demographic characteristics, disease history, medication use history, and stress test vital signs. We developed an Ensembling-based predictive model using 13 attributes that were selected based on their clinical importance, Multiple Linear Regression, and Information Gain Ranking methods. The negative effect of the imbalance class of the constructed model was handled by Synthetic Minority Oversampling Technique (SMOTE). The overall performance of the predictive model classifier was improved by the Ensemble machine learning approach using the Vote method with three Decision Trees (Naïve Bayes Tree, Random Forest, and Logistic Model Tree) and achieved high accuracy of prediction (AUC = 0.92). The study shows the potential of ensembling and SMOTE approaches for predicting incident diabetes using cardiorespiratory fitness data.

  11. Predicting diabetes mellitus using SMOTE and ensemble machine learning approach: The Henry Ford ExercIse Testing (FIT project.

    Directory of Open Access Journals (Sweden)

    Manal Alghamdi

    Full Text Available Machine learning is becoming a popular and important approach in the field of medical research. In this study, we investigate the relative performance of various machine learning methods such as Decision Tree, Naïve Bayes, Logistic Regression, Logistic Model Tree and Random Forests for predicting incident diabetes using medical records of cardiorespiratory fitness. In addition, we apply different techniques to uncover potential predictors of diabetes. This FIT project study used data of 32,555 patients who are free of any known coronary artery disease or heart failure who underwent clinician-referred exercise treadmill stress testing at Henry Ford Health Systems between 1991 and 2009 and had a complete 5-year follow-up. At the completion of the fifth year, 5,099 of those patients have developed diabetes. The dataset contained 62 attributes classified into four categories: demographic characteristics, disease history, medication use history, and stress test vital signs. We developed an Ensembling-based predictive model using 13 attributes that were selected based on their clinical importance, Multiple Linear Regression, and Information Gain Ranking methods. The negative effect of the imbalance class of the constructed model was handled by Synthetic Minority Oversampling Technique (SMOTE. The overall performance of the predictive model classifier was improved by the Ensemble machine learning approach using the Vote method with three Decision Trees (Naïve Bayes Tree, Random Forest, and Logistic Model Tree and achieved high accuracy of prediction (AUC = 0.92. The study shows the potential of ensembling and SMOTE approaches for predicting incident diabetes using cardiorespiratory fitness data.

  12. Testing a cue outside the training context increases attention to the contexts and impairs performance in human predictive learning.

    Science.gov (United States)

    Aristizabal, José A; Ramos-Álvarez, Manuel M; Callejas-Aguilera, José E; Rosas, Juan M

    2017-12-01

    One experiment in human predictive learning explored the impact of a context change on attention to contexts and predictive ratings controlled by the cue. In Context A: cue X was paired with an outcome four times, while cue Y was presented without an outcome four times in Context B:. In both contexts filler cues were presented without the outcome. During the test, target cues X and Y were presented either in the context where they were trained, or in the alternative context. With the context change expectation of the outcome X, expressed as predictive ratings, decreased in the presence of X and increased in the presence of Y. Looking at the contexts, expressed as a percentage of the overall gaze dwell time on a trial, was high across the four training trials, and increased with the context change. Results suggest that the presentation of unexpected information leads to increases in attention to contextual cues. Implications for contextual control of behavior are discussed. Copyright © 2017 Elsevier B.V. All rights reserved.

  13. Learning Python testing

    CERN Document Server

    Arbuckle, Daniel

    2014-01-01

    This book is ideal if you want to learn about the testing disciplines and automated testing tools from a hands-on, conversational guide. You should already know Python and be comfortable with Python 3.

  14. Comparison of machine learning techniques to predict all-cause mortality using fitness data: the Henry ford exercIse testing (FIT) project.

    Science.gov (United States)

    Sakr, Sherif; Elshawi, Radwa; Ahmed, Amjad M; Qureshi, Waqas T; Brawner, Clinton A; Keteyian, Steven J; Blaha, Michael J; Al-Mallah, Mouaz H

    2017-12-19

    Prior studies have demonstrated that cardiorespiratory fitness (CRF) is a strong marker of cardiovascular health. Machine learning (ML) can enhance the prediction of outcomes through classification techniques that classify the data into predetermined categories. The aim of this study is to present an evaluation and comparison of how machine learning techniques can be applied on medical records of cardiorespiratory fitness and how the various techniques differ in terms of capabilities of predicting medical outcomes (e.g. mortality). We use data of 34,212 patients free of known coronary artery disease or heart failure who underwent clinician-referred exercise treadmill stress testing at Henry Ford Health Systems Between 1991 and 2009 and had a complete 10-year follow-up. Seven machine learning classification techniques were evaluated: Decision Tree (DT), Support Vector Machine (SVM), Artificial Neural Networks (ANN), Naïve Bayesian Classifier (BC), Bayesian Network (BN), K-Nearest Neighbor (KNN) and Random Forest (RF). In order to handle the imbalanced dataset used, the Synthetic Minority Over-Sampling Technique (SMOTE) is used. Two set of experiments have been conducted with and without the SMOTE sampling technique. On average over different evaluation metrics, SVM Classifier has shown the lowest performance while other models like BN, BC and DT performed better. The RF classifier has shown the best performance (AUC = 0.97) among all models trained using the SMOTE sampling. The results show that various ML techniques can significantly vary in terms of its performance for the different evaluation metrics. It is also not necessarily that the more complex the ML model, the more prediction accuracy can be achieved. The prediction performance of all models trained with SMOTE is much better than the performance of models trained without SMOTE. The study shows the potential of machine learning methods for predicting all-cause mortality using cardiorespiratory fitness

  15. Signed reward prediction errors drive declarative learning.

    Directory of Open Access Journals (Sweden)

    Esther De Loof

    Full Text Available Reward prediction errors (RPEs are thought to drive learning. This has been established in procedural learning (e.g., classical and operant conditioning. However, empirical evidence on whether RPEs drive declarative learning-a quintessentially human form of learning-remains surprisingly absent. We therefore coupled RPEs to the acquisition of Dutch-Swahili word pairs in a declarative learning paradigm. Signed RPEs (SRPEs; "better-than-expected" signals during declarative learning improved recognition in a follow-up test, with increasingly positive RPEs leading to better recognition. In addition, classic declarative memory mechanisms such as time-on-task failed to explain recognition performance. The beneficial effect of SRPEs on recognition was subsequently affirmed in a replication study with visual stimuli.

  16. Signed reward prediction errors drive declarative learning.

    Science.gov (United States)

    De Loof, Esther; Ergo, Kate; Naert, Lien; Janssens, Clio; Talsma, Durk; Van Opstal, Filip; Verguts, Tom

    2018-01-01

    Reward prediction errors (RPEs) are thought to drive learning. This has been established in procedural learning (e.g., classical and operant conditioning). However, empirical evidence on whether RPEs drive declarative learning-a quintessentially human form of learning-remains surprisingly absent. We therefore coupled RPEs to the acquisition of Dutch-Swahili word pairs in a declarative learning paradigm. Signed RPEs (SRPEs; "better-than-expected" signals) during declarative learning improved recognition in a follow-up test, with increasingly positive RPEs leading to better recognition. In addition, classic declarative memory mechanisms such as time-on-task failed to explain recognition performance. The beneficial effect of SRPEs on recognition was subsequently affirmed in a replication study with visual stimuli.

  17. Learning to Predict Chemical Reactions

    Science.gov (United States)

    Kayala, Matthew A.; Azencott, Chloé-Agathe; Chen, Jonathan H.

    2011-01-01

    Being able to predict the course of arbitrary chemical reactions is essential to the theory and applications of organic chemistry. Approaches to the reaction prediction problems can be organized around three poles corresponding to: (1) physical laws; (2) rule-based expert systems; and (3) inductive machine learning. Previous approaches at these poles respectively are not high-throughput, are not generalizable or scalable, or lack sufficient data and structure to be implemented. We propose a new approach to reaction prediction utilizing elements from each pole. Using a physically inspired conceptualization, we describe single mechanistic reactions as interactions between coarse approximations of molecular orbitals (MOs) and use topological and physicochemical attributes as descriptors. Using an existing rule-based system (Reaction Explorer), we derive a restricted chemistry dataset consisting of 1630 full multi-step reactions with 2358 distinct starting materials and intermediates, associated with 2989 productive mechanistic steps and 6.14 million unproductive mechanistic steps. And from machine learning, we pose identifying productive mechanistic steps as a statistical ranking, information retrieval, problem: given a set of reactants and a description of conditions, learn a ranking model over potential filled-to-unfilled MO interactions such that the top ranked mechanistic steps yield the major products. The machine learning implementation follows a two-stage approach, in which we first train atom level reactivity filters to prune 94.00% of non-productive reactions with a 0.01% error rate. Then, we train an ensemble of ranking models on pairs of interacting MOs to learn a relative productivity function over mechanistic steps in a given system. Without the use of explicit transformation patterns, the ensemble perfectly ranks the productive mechanism at the top 89.05% of the time, rising to 99.86% of the time when the top four are considered. Furthermore, the system

  18. Learning to predict chemical reactions.

    Science.gov (United States)

    Kayala, Matthew A; Azencott, Chloé-Agathe; Chen, Jonathan H; Baldi, Pierre

    2011-09-26

    Being able to predict the course of arbitrary chemical reactions is essential to the theory and applications of organic chemistry. Approaches to the reaction prediction problems can be organized around three poles corresponding to: (1) physical laws; (2) rule-based expert systems; and (3) inductive machine learning. Previous approaches at these poles, respectively, are not high throughput, are not generalizable or scalable, and lack sufficient data and structure to be implemented. We propose a new approach to reaction prediction utilizing elements from each pole. Using a physically inspired conceptualization, we describe single mechanistic reactions as interactions between coarse approximations of molecular orbitals (MOs) and use topological and physicochemical attributes as descriptors. Using an existing rule-based system (Reaction Explorer), we derive a restricted chemistry data set consisting of 1630 full multistep reactions with 2358 distinct starting materials and intermediates, associated with 2989 productive mechanistic steps and 6.14 million unproductive mechanistic steps. And from machine learning, we pose identifying productive mechanistic steps as a statistical ranking, information retrieval problem: given a set of reactants and a description of conditions, learn a ranking model over potential filled-to-unfilled MO interactions such that the top-ranked mechanistic steps yield the major products. The machine learning implementation follows a two-stage approach, in which we first train atom level reactivity filters to prune 94.00% of nonproductive reactions with a 0.01% error rate. Then, we train an ensemble of ranking models on pairs of interacting MOs to learn a relative productivity function over mechanistic steps in a given system. Without the use of explicit transformation patterns, the ensemble perfectly ranks the productive mechanism at the top 89.05% of the time, rising to 99.86% of the time when the top four are considered. Furthermore, the system

  19. A causal link between prediction errors, dopamine neurons and learning.

    Science.gov (United States)

    Steinberg, Elizabeth E; Keiflin, Ronald; Boivin, Josiah R; Witten, Ilana B; Deisseroth, Karl; Janak, Patricia H

    2013-07-01

    Situations in which rewards are unexpectedly obtained or withheld represent opportunities for new learning. Often, this learning includes identifying cues that predict reward availability. Unexpected rewards strongly activate midbrain dopamine neurons. This phasic signal is proposed to support learning about antecedent cues by signaling discrepancies between actual and expected outcomes, termed a reward prediction error. However, it is unknown whether dopamine neuron prediction error signaling and cue-reward learning are causally linked. To test this hypothesis, we manipulated dopamine neuron activity in rats in two behavioral procedures, associative blocking and extinction, that illustrate the essential function of prediction errors in learning. We observed that optogenetic activation of dopamine neurons concurrent with reward delivery, mimicking a prediction error, was sufficient to cause long-lasting increases in cue-elicited reward-seeking behavior. Our findings establish a causal role for temporally precise dopamine neuron signaling in cue-reward learning, bridging a critical gap between experimental evidence and influential theoretical frameworks.

  20. Learning Predictive Statistics: Strategies and Brain Mechanisms.

    Science.gov (United States)

    Wang, Rui; Shen, Yuan; Tino, Peter; Welchman, Andrew E; Kourtzi, Zoe

    2017-08-30

    When immersed in a new environment, we are challenged to decipher initially incomprehensible streams of sensory information. However, quite rapidly, the brain finds structure and meaning in these incoming signals, helping us to predict and prepare ourselves for future actions. This skill relies on extracting the statistics of event streams in the environment that contain regularities of variable complexity from simple repetitive patterns to complex probabilistic combinations. Here, we test the brain mechanisms that mediate our ability to adapt to the environment's statistics and predict upcoming events. By combining behavioral training and multisession fMRI in human participants (male and female), we track the corticostriatal mechanisms that mediate learning of temporal sequences as they change in structure complexity. We show that learning of predictive structures relates to individual decision strategy; that is, selecting the most probable outcome in a given context (maximizing) versus matching the exact sequence statistics. These strategies engage distinct human brain regions: maximizing engages dorsolateral prefrontal, cingulate, sensory-motor regions, and basal ganglia (dorsal caudate, putamen), whereas matching engages occipitotemporal regions (including the hippocampus) and basal ganglia (ventral caudate). Our findings provide evidence for distinct corticostriatal mechanisms that facilitate our ability to extract behaviorally relevant statistics to make predictions. SIGNIFICANCE STATEMENT Making predictions about future events relies on interpreting streams of information that may initially appear incomprehensible. Past work has studied how humans identify repetitive patterns and associative pairings. However, the natural environment contains regularities that vary in complexity from simple repetition to complex probabilistic combinations. Here, we combine behavior and multisession fMRI to track the brain mechanisms that mediate our ability to adapt to

  1. Using Tests as Learning Opportunities.

    Science.gov (United States)

    Foos, Paul W.; Fisher, Ronald P.

    1988-01-01

    A study involving 105 undergraduates assessed the value of testing as a means of increasing, rather than simply monitoring, learning. Results indicate that fill-in-the-blank and items requiring student inferences were more effective, respectively, than multiple-choice tests and verbatim items in furthering student learning. (TJH)

  2. Critical evidence for the prediction error theory in associative learning.

    Science.gov (United States)

    Terao, Kanta; Matsumoto, Yukihisa; Mizunami, Makoto

    2015-03-10

    In associative learning in mammals, it is widely accepted that the discrepancy, or error, between actual and predicted reward determines whether learning occurs. Complete evidence for the prediction error theory, however, has not been obtained in any learning systems: Prediction error theory stems from the finding of a blocking phenomenon, but blocking can also be accounted for by other theories, such as the attentional theory. We demonstrated blocking in classical conditioning in crickets and obtained evidence to reject the attentional theory. To obtain further evidence supporting the prediction error theory and rejecting alternative theories, we constructed a neural model to match the prediction error theory, by modifying our previous model of learning in crickets, and we tested a prediction from the model: the model predicts that pharmacological intervention of octopaminergic transmission during appetitive conditioning impairs learning but not formation of reward prediction itself, and it thus predicts no learning in subsequent training. We observed such an "auto-blocking", which could be accounted for by the prediction error theory but not by other competitive theories to account for blocking. This study unambiguously demonstrates validity of the prediction error theory in associative learning.

  3. Learning software testing with Test Studio

    CERN Document Server

    Madi, Rawane

    2013-01-01

    Learning Software Testing with Test Studio is a practical, hands-on guide that will help you get started with Test Studio to design your automated solution and tests. All through the book, there are best practices and tips and tricks inside Test Studio which can be employed to improve your solution just like an experienced QA.If you are a beginner or a professional QA who is seeking a fast, clear, and direct to the point start in automated software testing inside Test Studio, this book is for you. You should be familiar with the .NET framework, mainly Visual Studio, C#, and SQL, as the book's

  4. Pretest Predictions for Ventilation Tests

    International Nuclear Information System (INIS)

    Y. Sun; H. Yang; H.N. Kalia

    2007-01-01

    The objective of this calculation is to predict the temperatures of the ventilating air, waste package surface, concrete pipe walls, and insulation that will be developed during the ventilation tests involving various test conditions. The results will be used as input to the following three areas: (1) Decisions regarding testing set-up and performance. (2) Assessing how best to scale the test phenomena measured. (3) Validating numerical approach for modeling continuous ventilation. The scope of the calculation is to identify the physical mechanisms and parameters related to thermal response in the ventilation tests, and develop and describe numerical methods that can be used to calculate the effects of continuous ventilation. Sensitivity studies to assess the impact of variation of linear power densities (linear heat loads) and ventilation air flow rates are included. The calculation is limited to thermal effect only

  5. Learning Android application testing

    CERN Document Server

    Blundell, Paul

    2015-01-01

    If you are an Android developer looking to test your applications or optimize your application development process, then this book is for you. No previous experience in application testing is required.

  6. Using Machine Learning to Predict Student Performance

    OpenAIRE

    Pojon, Murat

    2017-01-01

    This thesis examines the application of machine learning algorithms to predict whether a student will be successful or not. The specific focus of the thesis is the comparison of machine learning methods and feature engineering techniques in terms of how much they improve the prediction performance. Three different machine learning methods were used in this thesis. They are linear regression, decision trees, and naïve Bayes classification. Feature engineering, the process of modification ...

  7. Predicting Student Success from the "LASSI for Learning Online" (LLO)

    Science.gov (United States)

    Carson, Andrew D.

    2011-01-01

    This study tested the degree to which subscales of the "LASSI for Learning Online" (LLO) (Weinstein & Palmer, 2006), a measure of learning strategies and study skills, predict student success in the form of passing grades, using a combination of large training (N = 4,409) and cross-validation (N = 3,203) samples. Discriminant function analysis…

  8. Predicting Increased Blood Pressure Using Machine Learning

    Science.gov (United States)

    Golino, Hudson Fernandes; Amaral, Liliany Souza de Brito; Duarte, Stenio Fernando Pimentel; Soares, Telma de Jesus; dos Reis, Luciana Araujo

    2014-01-01

    The present study investigates the prediction of increased blood pressure by body mass index (BMI), waist (WC) and hip circumference (HC), and waist hip ratio (WHR) using a machine learning technique named classification tree. Data were collected from 400 college students (56.3% women) from 16 to 63 years old. Fifteen trees were calculated in the training group for each sex, using different numbers and combinations of predictors. The result shows that for women BMI, WC, and WHR are the combination that produces the best prediction, since it has the lowest deviance (87.42), misclassification (.19), and the higher pseudo R 2 (.43). This model presented a sensitivity of 80.86% and specificity of 81.22% in the training set and, respectively, 45.65% and 65.15% in the test sample. For men BMI, WC, HC, and WHC showed the best prediction with the lowest deviance (57.25), misclassification (.16), and the higher pseudo R 2 (.46). This model had a sensitivity of 72% and specificity of 86.25% in the training set and, respectively, 58.38% and 69.70% in the test set. Finally, the result from the classification tree analysis was compared with traditional logistic regression, indicating that the former outperformed the latter in terms of predictive power. PMID:24669313

  9. Predicting increased blood pressure using machine learning.

    Science.gov (United States)

    Golino, Hudson Fernandes; Amaral, Liliany Souza de Brito; Duarte, Stenio Fernando Pimentel; Gomes, Cristiano Mauro Assis; Soares, Telma de Jesus; Dos Reis, Luciana Araujo; Santos, Joselito

    2014-01-01

    The present study investigates the prediction of increased blood pressure by body mass index (BMI), waist (WC) and hip circumference (HC), and waist hip ratio (WHR) using a machine learning technique named classification tree. Data were collected from 400 college students (56.3% women) from 16 to 63 years old. Fifteen trees were calculated in the training group for each sex, using different numbers and combinations of predictors. The result shows that for women BMI, WC, and WHR are the combination that produces the best prediction, since it has the lowest deviance (87.42), misclassification (.19), and the higher pseudo R (2) (.43). This model presented a sensitivity of 80.86% and specificity of 81.22% in the training set and, respectively, 45.65% and 65.15% in the test sample. For men BMI, WC, HC, and WHC showed the best prediction with the lowest deviance (57.25), misclassification (.16), and the higher pseudo R (2) (.46). This model had a sensitivity of 72% and specificity of 86.25% in the training set and, respectively, 58.38% and 69.70% in the test set. Finally, the result from the classification tree analysis was compared with traditional logistic regression, indicating that the former outperformed the latter in terms of predictive power.

  10. Predicting Increased Blood Pressure Using Machine Learning

    Directory of Open Access Journals (Sweden)

    Hudson Fernandes Golino

    2014-01-01

    Full Text Available The present study investigates the prediction of increased blood pressure by body mass index (BMI, waist (WC and hip circumference (HC, and waist hip ratio (WHR using a machine learning technique named classification tree. Data were collected from 400 college students (56.3% women from 16 to 63 years old. Fifteen trees were calculated in the training group for each sex, using different numbers and combinations of predictors. The result shows that for women BMI, WC, and WHR are the combination that produces the best prediction, since it has the lowest deviance (87.42, misclassification (.19, and the higher pseudo R2 (.43. This model presented a sensitivity of 80.86% and specificity of 81.22% in the training set and, respectively, 45.65% and 65.15% in the test sample. For men BMI, WC, HC, and WHC showed the best prediction with the lowest deviance (57.25, misclassification (.16, and the higher pseudo R2 (.46. This model had a sensitivity of 72% and specificity of 86.25% in the training set and, respectively, 58.38% and 69.70% in the test set. Finally, the result from the classification tree analysis was compared with traditional logistic regression, indicating that the former outperformed the latter in terms of predictive power.

  11. Video Scene Parsing with Predictive Feature Learning

    OpenAIRE

    Jin, Xiaojie; Li, Xin; Xiao, Huaxin; Shen, Xiaohui; Lin, Zhe; Yang, Jimei; Chen, Yunpeng; Dong, Jian; Liu, Luoqi; Jie, Zequn; Feng, Jiashi; Yan, Shuicheng

    2016-01-01

    In this work, we address the challenging video scene parsing problem by developing effective representation learning methods given limited parsing annotations. In particular, we contribute two novel methods that constitute a unified parsing framework. (1) \\textbf{Predictive feature learning}} from nearly unlimited unlabeled video data. Different from existing methods learning features from single frame parsing, we learn spatiotemporal discriminative features by enforcing a parsing network to ...

  12. Predicting Process Behaviour using Deep Learning

    OpenAIRE

    Evermann, Joerg; Rehse, Jana-Rebecca; Fettke, Peter

    2016-01-01

    Predicting business process behaviour is an important aspect of business process management. Motivated by research in natural language processing, this paper describes an application of deep learning with recurrent neural networks to the problem of predicting the next event in a business process. This is both a novel method in process prediction, which has largely relied on explicit process models, and also a novel application of deep learning methods. The approach is evaluated on two real da...

  13. Predicting radiotherapy outcomes using statistical learning techniques

    International Nuclear Information System (INIS)

    El Naqa, Issam; Bradley, Jeffrey D; Deasy, Joseph O; Lindsay, Patricia E; Hope, Andrew J

    2009-01-01

    Radiotherapy outcomes are determined by complex interactions between treatment, anatomical and patient-related variables. A common obstacle to building maximally predictive outcome models for clinical practice is the failure to capture potential complexity of heterogeneous variable interactions and applicability beyond institutional data. We describe a statistical learning methodology that can automatically screen for nonlinear relations among prognostic variables and generalize to unseen data before. In this work, several types of linear and nonlinear kernels to generate interaction terms and approximate the treatment-response function are evaluated. Examples of institutional datasets of esophagitis, pneumonitis and xerostomia endpoints were used. Furthermore, an independent RTOG dataset was used for 'generalizabilty' validation. We formulated the discrimination between risk groups as a supervised learning problem. The distribution of patient groups was initially analyzed using principle components analysis (PCA) to uncover potential nonlinear behavior. The performance of the different methods was evaluated using bivariate correlations and actuarial analysis. Over-fitting was controlled via cross-validation resampling. Our results suggest that a modified support vector machine (SVM) kernel method provided superior performance on leave-one-out testing compared to logistic regression and neural networks in cases where the data exhibited nonlinear behavior on PCA. For instance, in prediction of esophagitis and pneumonitis endpoints, which exhibited nonlinear behavior on PCA, the method provided 21% and 60% improvements, respectively. Furthermore, evaluation on the independent pneumonitis RTOG dataset demonstrated good generalizabilty beyond institutional data in contrast with other models. This indicates that the prediction of treatment response can be improved by utilizing nonlinear kernel methods for discovering important nonlinear interactions among model

  14. Recent Advances in Predictive (Machine) Learning

    Energy Technology Data Exchange (ETDEWEB)

    Friedman, J

    2004-01-24

    Prediction involves estimating the unknown value of an attribute of a system under study given the values of other measured attributes. In prediction (machine) learning the prediction rule is derived from data consisting of previously solved cases. Most methods for predictive learning were originated many years ago at the dawn of the computer age. Recently two new techniques have emerged that have revitalized the field. These are support vector machines and boosted decision trees. This paper provides an introduction to these two new methods tracing their respective ancestral roots to standard kernel methods and ordinary decision trees.

  15. Signed reward prediction errors drive declarative learning

    NARCIS (Netherlands)

    De Loof, E.; Ergo, K.; Naert, L.; Janssens, C.; Talsma, D.; van Opstal, F.; Verguts, T.

    2018-01-01

    Reward prediction errors (RPEs) are thought to drive learning. This has been established in procedural learning (e.g., classical and operant conditioning). However, empirical evidence on whether RPEs drive declarative learning–a quintessentially human form of learning–remains surprisingly absent. We

  16. Predicting Learned Helplessness Based on Personality

    Science.gov (United States)

    Maadikhah, Elham; Erfani, Nasrollah

    2014-01-01

    Learned helplessness as a negative motivational state can latently underlie repeated failures and create negative feelings toward the education as well as depression in students and other members of a society. The purpose of this paper is to predict learned helplessness based on students' personality traits. The research is a predictive…

  17. Randomized Prediction Games for Adversarial Machine Learning.

    Science.gov (United States)

    Rota Bulo, Samuel; Biggio, Battista; Pillai, Ignazio; Pelillo, Marcello; Roli, Fabio

    In spam and malware detection, attackers exploit randomization to obfuscate malicious data and increase their chances of evading detection at test time, e.g., malware code is typically obfuscated using random strings or byte sequences to hide known exploits. Interestingly, randomization has also been proposed to improve security of learning algorithms against evasion attacks, as it results in hiding information about the classifier to the attacker. Recent work has proposed game-theoretical formulations to learn secure classifiers, by simulating different evasion attacks and modifying the classification function accordingly. However, both the classification function and the simulated data manipulations have been modeled in a deterministic manner, without accounting for any form of randomization. In this paper, we overcome this limitation by proposing a randomized prediction game, namely, a noncooperative game-theoretic formulation in which the classifier and the attacker make randomized strategy selections according to some probability distribution defined over the respective strategy set. We show that our approach allows one to improve the tradeoff between attack detection and false alarms with respect to the state-of-the-art secure classifiers, even against attacks that are different from those hypothesized during design, on application examples including handwritten digit recognition, spam, and malware detection.In spam and malware detection, attackers exploit randomization to obfuscate malicious data and increase their chances of evading detection at test time, e.g., malware code is typically obfuscated using random strings or byte sequences to hide known exploits. Interestingly, randomization has also been proposed to improve security of learning algorithms against evasion attacks, as it results in hiding information about the classifier to the attacker. Recent work has proposed game-theoretical formulations to learn secure classifiers, by simulating different

  18. Predicting breast screening attendance using machine learning techniques.

    Science.gov (United States)

    Baskaran, Vikraman; Guergachi, Aziz; Bali, Rajeev K; Naguib, Raouf N G

    2011-03-01

    Machine learning-based prediction has been effectively applied for many healthcare applications. Predicting breast screening attendance using machine learning (prior to the actual mammogram) is a new field. This paper presents new predictor attributes for such an algorithm. It describes a new hybrid algorithm that relies on back-propagation and radial basis function-based neural networks for prediction. The algorithm has been developed in an open source-based environment. The algorithm was tested on a 13-year dataset (1995-2008). This paper compares the algorithm and validates its accuracy and efficiency with different platforms. Nearly 80% accuracy and 88% positive predictive value and sensitivity were recorded for the algorithm. The results were encouraging; 40-50% of negative predictive value and specificity warrant further work. Preliminary results were promising and provided ample amount of reasons for testing the algorithm on a larger scale.

  19. Testing a machine-learning algorithm to predict the persistence and severity of major depressive disorder from baseline self-reports.

    Science.gov (United States)

    Kessler, R C; van Loo, H M; Wardenaar, K J; Bossarte, R M; Brenner, L A; Cai, T; Ebert, D D; Hwang, I; Li, J; de Jonge, P; Nierenberg, A A; Petukhova, M V; Rosellini, A J; Sampson, N A; Schoevers, R A; Wilcox, M A; Zaslavsky, A M

    2016-10-01

    Heterogeneity of major depressive disorder (MDD) illness course complicates clinical decision-making. Although efforts to use symptom profiles or biomarkers to develop clinically useful prognostic subtypes have had limited success, a recent report showed that machine-learning (ML) models developed from self-reports about incident episode characteristics and comorbidities among respondents with lifetime MDD in the World Health Organization World Mental Health (WMH) Surveys predicted MDD persistence, chronicity and severity with good accuracy. We report results of model validation in an independent prospective national household sample of 1056 respondents with lifetime MDD at baseline. The WMH ML models were applied to these baseline data to generate predicted outcome scores that were compared with observed scores assessed 10-12 years after baseline. ML model prediction accuracy was also compared with that of conventional logistic regression models. Area under the receiver operating characteristic curve based on ML (0.63 for high chronicity and 0.71-0.76 for the other prospective outcomes) was consistently higher than for the logistic models (0.62-0.70) despite the latter models including more predictors. A total of 34.6-38.1% of respondents with subsequent high persistence chronicity and 40.8-55.8% with the severity indicators were in the top 20% of the baseline ML-predicted risk distribution, while only 0.9% of respondents with subsequent hospitalizations and 1.5% with suicide attempts were in the lowest 20% of the ML-predicted risk distribution. These results confirm that clinically useful MDD risk-stratification models can be generated from baseline patient self-reports and that ML methods improve on conventional methods in developing such models.

  20. Predikce poruch učení pomocí testu komplexní imitace pohybu Prediction of learning difficulties with the test of complex imitation of movement

    Directory of Open Access Journals (Sweden)

    Martina Ozbič

    2008-05-01

    can lead to faster intervention resulting in the progress of children with DCD in their movement abilities. This research has shown that on the basis of twenty tasks of the Bergès-Lézine's test of the complex imitation of movement/gestures, we can predict which children have some learning difficulties and which do not. Particularly we wish to emphasize three tasks (12, 17 and 20 where children had to cross the vertical midline of their bodies. These three tasks involve bilateral coordination. Children with DCD signs face problems in spatial orientation and in complex imitation of movement/gestures. On the basis of great differences, found in tasks where pupils had to cross the vertical midline of their bodies and rotate their palms, children can be classified into two groups (with and without motor and learning difficulties.

  1. Predicting Virtual Learning Environment Adoption

    DEFF Research Database (Denmark)

    Penjor, Sonam; Zander, Pär-Ola Mikael

    2016-01-01

    This study investigates the significance of Rogers’ Diffusion of Innovations (DOI) theory with regard to the use of a Virtual Learning Environment (VLE) at the Royal University of Bhutan (RUB). The focus is on different adoption types and characteristics of users. Rogers’ DOI theory is applied...... to investigate the influence of five predictors (relative advantage, complexity, compatibility, trialability and observability) and their significance in the perception of academic staff at the RUB in relation to the probability of VLE adoption. These predictors are attributes of the VLE that determine the rate...... of adoption by various adopter group memberships (Innovators, Early Adopters, Early Majority, Late Majority, Laggards). Descriptive statistics and regression analysis were deployed to analyse adopter group memberships and predictor significance in VLE adoption and use. The results revealed varying attitudes...

  2. Huntington's disease : Psychological aspects of predictive testing

    NARCIS (Netherlands)

    Timman, Reinier

    2005-01-01

    Predictive testing for Huntington's disease appears to have long lasting psychological effects. The predictive test for Huntington's disease (HD), a hereditary disease of the nervous system, was introduced in the Netherlands in the late eighties. As adverse consequences of the test were

  3. Do Judgments of Learning Predict Automatic Influences of Memory?

    Science.gov (United States)

    Undorf, Monika; Böhm, Simon; Cüpper, Lutz

    2016-01-01

    Current memory theories generally assume that memory performance reflects both recollection and automatic influences of memory. Research on people's predictions about the likelihood of remembering recently studied information on a memory test, that is, on judgments of learning (JOLs), suggests that both magnitude and resolution of JOLs are linked…

  4. Machine Learning Principles Can Improve Hip Fracture Prediction

    DEFF Research Database (Denmark)

    Kruse, Christian; Eiken, Pia; Vestergaard, Peter

    2017-01-01

    Apply machine learning principles to predict hip fractures and estimate predictor importance in Dual-energy X-ray absorptiometry (DXA)-scanned men and women. Dual-energy X-ray absorptiometry data from two Danish regions between 1996 and 2006 were combined with national Danish patient data.......89 [0.82; 0.95], but with poor calibration in higher probabilities. A ten predictor subset (BMD, biochemical cholesterol and liver function tests, penicillin use and osteoarthritis diagnoses) achieved a test AUC of 0.86 [0.78; 0.94] using an “xgbTree” model. Machine learning can improve hip fracture...... prediction beyond logistic regression using ensemble models. Compiling data from international cohorts of longer follow-up and performing similar machine learning procedures has the potential to further improve discrimination and calibration....

  5. Improving orbit prediction accuracy through supervised machine learning

    Science.gov (United States)

    Peng, Hao; Bai, Xiaoli

    2018-05-01

    Due to the lack of information such as the space environment condition and resident space objects' (RSOs') body characteristics, current orbit predictions that are solely grounded on physics-based models may fail to achieve required accuracy for collision avoidance and have led to satellite collisions already. This paper presents a methodology to predict RSOs' trajectories with higher accuracy than that of the current methods. Inspired by the machine learning (ML) theory through which the models are learned based on large amounts of observed data and the prediction is conducted without explicitly modeling space objects and space environment, the proposed ML approach integrates physics-based orbit prediction algorithms with a learning-based process that focuses on reducing the prediction errors. Using a simulation-based space catalog environment as the test bed, the paper demonstrates three types of generalization capability for the proposed ML approach: (1) the ML model can be used to improve the same RSO's orbit information that is not available during the learning process but shares the same time interval as the training data; (2) the ML model can be used to improve predictions of the same RSO at future epochs; and (3) the ML model based on a RSO can be applied to other RSOs that share some common features.

  6. Predicting occupational personality test scores.

    Science.gov (United States)

    Furnham, A; Drakeley, R

    2000-01-01

    The relationship between students' actual test scores and their self-estimated scores on the Hogan Personality Inventory (HPI; R. Hogan & J. Hogan, 1992), an omnibus personality questionnaire, was examined. Despite being given descriptive statistics and explanations of each of the dimensions measured, the students tended to overestimate their scores; yet all correlations between actual and estimated scores were positive and significant. Correlations between self-estimates and actual test scores were highest for sociability, ambition, and adjustment (r = .62 to r = .67). The results are discussed in terms of employers' use and abuse of personality assessment for job recruitment.

  7. Strength of Temporal White Matter Pathways Predicts Semantic Learning.

    Science.gov (United States)

    Ripollés, Pablo; Biel, Davina; Peñaloza, Claudia; Kaufmann, Jörn; Marco-Pallarés, Josep; Noesselt, Toemme; Rodríguez-Fornells, Antoni

    2017-11-15

    Learning the associations between words and meanings is a fundamental human ability. Although the language network is cortically well defined, the role of the white matter pathways supporting novel word-to-meaning mappings remains unclear. Here, by using contextual and cross-situational word learning, we tested whether learning the meaning of a new word is related to the integrity of the language-related white matter pathways in 40 adults (18 women). The arcuate, uncinate, inferior-fronto-occipital and inferior-longitudinal fasciculi were virtually dissected using manual and automatic deterministic fiber tracking. Critically, the automatic method allowed assessing the white matter microstructure along the tract. Results demonstrate that the microstructural properties of the left inferior-longitudinal fasciculus predict contextual learning, whereas the left uncinate was associated with cross-situational learning. In addition, we identified regions of special importance within these pathways: the posterior middle temporal gyrus, thought to serve as a lexical interface and specifically related to contextual learning; the anterior temporal lobe, known to be an amodal hub for semantic processing and related to cross-situational learning; and the white matter near the hippocampus, a structure fundamental for the initial stages of new-word learning and, remarkably, related to both types of word learning. No significant associations were found for the inferior-fronto-occipital fasciculus or the arcuate. While previous results suggest that learning new phonological word forms is mediated by the arcuate fasciculus, these findings show that the temporal pathways are the crucial neural substrate supporting one of the most striking human abilities: our capacity to identify correct associations between words and meanings under referential indeterminacy. SIGNIFICANCE STATEMENT The language-processing network is cortically (i.e., gray matter) well defined. However, the role of the

  8. Machine Learning Methods to Predict Diabetes Complications.

    Science.gov (United States)

    Dagliati, Arianna; Marini, Simone; Sacchi, Lucia; Cogni, Giulia; Teliti, Marsida; Tibollo, Valentina; De Cata, Pasquale; Chiovato, Luca; Bellazzi, Riccardo

    2018-03-01

    One of the areas where Artificial Intelligence is having more impact is machine learning, which develops algorithms able to learn patterns and decision rules from data. Machine learning algorithms have been embedded into data mining pipelines, which can combine them with classical statistical strategies, to extract knowledge from data. Within the EU-funded MOSAIC project, a data mining pipeline has been used to derive a set of predictive models of type 2 diabetes mellitus (T2DM) complications based on electronic health record data of nearly one thousand patients. Such pipeline comprises clinical center profiling, predictive model targeting, predictive model construction and model validation. After having dealt with missing data by means of random forest (RF) and having applied suitable strategies to handle class imbalance, we have used Logistic Regression with stepwise feature selection to predict the onset of retinopathy, neuropathy, or nephropathy, at different time scenarios, at 3, 5, and 7 years from the first visit at the Hospital Center for Diabetes (not from the diagnosis). Considered variables are gender, age, time from diagnosis, body mass index (BMI), glycated hemoglobin (HbA1c), hypertension, and smoking habit. Final models, tailored in accordance with the complications, provided an accuracy up to 0.838. Different variables were selected for each complication and time scenario, leading to specialized models easy to translate to the clinical practice.

  9. Auditory working memory predicts individual differences in absolute pitch learning.

    Science.gov (United States)

    Van Hedger, Stephen C; Heald, Shannon L M; Koch, Rachelle; Nusbaum, Howard C

    2015-07-01

    Absolute pitch (AP) is typically defined as the ability to label an isolated tone as a musical note in the absence of a reference tone. At first glance the acquisition of AP note categories seems like a perceptual learning task, since individuals must assign a category label to a stimulus based on a single perceptual dimension (pitch) while ignoring other perceptual dimensions (e.g., loudness, octave, instrument). AP, however, is rarely discussed in terms of domain-general perceptual learning mechanisms. This is because AP is typically assumed to depend on a critical period of development, in which early exposure to pitches and musical labels is thought to be necessary for the development of AP precluding the possibility of adult acquisition of AP. Despite this view of AP, several previous studies have found evidence that absolute pitch category learning is, to an extent, trainable in a post-critical period adult population, even if the performance typically achieved by this population is below the performance of a "true" AP possessor. The current studies attempt to understand the individual differences in learning to categorize notes using absolute pitch cues by testing a specific prediction regarding cognitive capacity related to categorization - to what extent does an individual's general auditory working memory capacity (WMC) predict the success of absolute pitch category acquisition. Since WMC has been shown to predict performance on a wide variety of other perceptual and category learning tasks, we predict that individuals with higher WMC should be better at learning absolute pitch note categories than individuals with lower WMC. Across two studies, we demonstrate that auditory WMC predicts the efficacy of learning absolute pitch note categories. These results suggest that a higher general auditory WMC might underlie the formation of absolute pitch categories for post-critical period adults. Implications for understanding the mechanisms that underlie the

  10. Applications of machine learning in cancer prediction and prognosis.

    Science.gov (United States)

    Cruz, Joseph A; Wishart, David S

    2007-02-11

    Machine learning is a branch of artificial intelligence that employs a variety of statistical, probabilistic and optimization techniques that allows computers to "learn" from past examples and to detect hard-to-discern patterns from large, noisy or complex data sets. This capability is particularly well-suited to medical applications, especially those that depend on complex proteomic and genomic measurements. As a result, machine learning is frequently used in cancer diagnosis and detection. More recently machine learning has been applied to cancer prognosis and prediction. This latter approach is particularly interesting as it is part of a growing trend towards personalized, predictive medicine. In assembling this review we conducted a broad survey of the different types of machine learning methods being used, the types of data being integrated and the performance of these methods in cancer prediction and prognosis. A number of trends are noted, including a growing dependence on protein biomarkers and microarray data, a strong bias towards applications in prostate and breast cancer, and a heavy reliance on "older" technologies such artificial neural networks (ANNs) instead of more recently developed or more easily interpretable machine learning methods. A number of published studies also appear to lack an appropriate level of validation or testing. Among the better designed and validated studies it is clear that machine learning methods can be used to substantially (15-25%) improve the accuracy of predicting cancer susceptibility, recurrence and mortality. At a more fundamental level, it is also evident that machine learning is also helping to improve our basic understanding of cancer development and progression.

  11. Machine Learning and Conflict Prediction: A Use Case

    Directory of Open Access Journals (Sweden)

    Chris Perry

    2013-10-01

    Full Text Available For at least the last two decades, the international community in general and the United Nations specifically have attempted to develop robust, accurate and effective conflict early warning system for conflict prevention. One potential and promising component of integrated early warning systems lies in the field of machine learning. This paper aims at giving conflict analysis a basic understanding of machine learning methodology as well as to test the feasibility and added value of such an approach. The paper finds that the selection of appropriate machine learning methodologies can offer substantial improvements in accuracy and performance. It also finds that even at this early stage in testing machine learning on conflict prediction, full models offer more predictive power than simply using a prior outbreak of violence as the leading indicator of current violence. This suggests that a refined data selection methodology combined with strategic use of machine learning algorithms could indeed offer a significant addition to the early warning toolkit. Finally, the paper suggests a number of steps moving forward to improve upon this initial test methodology.

  12. Machine learning methods for metabolic pathway prediction

    Directory of Open Access Journals (Sweden)

    Karp Peter D

    2010-01-01

    Full Text Available Abstract Background A key challenge in systems biology is the reconstruction of an organism's metabolic network from its genome sequence. One strategy for addressing this problem is to predict which metabolic pathways, from a reference database of known pathways, are present in the organism, based on the annotated genome of the organism. Results To quantitatively validate methods for pathway prediction, we developed a large "gold standard" dataset of 5,610 pathway instances known to be present or absent in curated metabolic pathway databases for six organisms. We defined a collection of 123 pathway features, whose information content we evaluated with respect to the gold standard. Feature data were used as input to an extensive collection of machine learning (ML methods, including naïve Bayes, decision trees, and logistic regression, together with feature selection and ensemble methods. We compared the ML methods to the previous PathoLogic algorithm for pathway prediction using the gold standard dataset. We found that ML-based prediction methods can match the performance of the PathoLogic algorithm. PathoLogic achieved an accuracy of 91% and an F-measure of 0.786. The ML-based prediction methods achieved accuracy as high as 91.2% and F-measure as high as 0.787. The ML-based methods output a probability for each predicted pathway, whereas PathoLogic does not, which provides more information to the user and facilitates filtering of predicted pathways. Conclusions ML methods for pathway prediction perform as well as existing methods, and have qualitative advantages in terms of extensibility, tunability, and explainability. More advanced prediction methods and/or more sophisticated input features may improve the performance of ML methods. However, pathway prediction performance appears to be limited largely by the ability to correctly match enzymes to the reactions they catalyze based on genome annotations.

  13. Machine learning methods for metabolic pathway prediction

    Science.gov (United States)

    2010-01-01

    Background A key challenge in systems biology is the reconstruction of an organism's metabolic network from its genome sequence. One strategy for addressing this problem is to predict which metabolic pathways, from a reference database of known pathways, are present in the organism, based on the annotated genome of the organism. Results To quantitatively validate methods for pathway prediction, we developed a large "gold standard" dataset of 5,610 pathway instances known to be present or absent in curated metabolic pathway databases for six organisms. We defined a collection of 123 pathway features, whose information content we evaluated with respect to the gold standard. Feature data were used as input to an extensive collection of machine learning (ML) methods, including naïve Bayes, decision trees, and logistic regression, together with feature selection and ensemble methods. We compared the ML methods to the previous PathoLogic algorithm for pathway prediction using the gold standard dataset. We found that ML-based prediction methods can match the performance of the PathoLogic algorithm. PathoLogic achieved an accuracy of 91% and an F-measure of 0.786. The ML-based prediction methods achieved accuracy as high as 91.2% and F-measure as high as 0.787. The ML-based methods output a probability for each predicted pathway, whereas PathoLogic does not, which provides more information to the user and facilitates filtering of predicted pathways. Conclusions ML methods for pathway prediction perform as well as existing methods, and have qualitative advantages in terms of extensibility, tunability, and explainability. More advanced prediction methods and/or more sophisticated input features may improve the performance of ML methods. However, pathway prediction performance appears to be limited largely by the ability to correctly match enzymes to the reactions they catalyze based on genome annotations. PMID:20064214

  14. Applications of Machine Learning in Cancer Prediction and Prognosis

    Directory of Open Access Journals (Sweden)

    Joseph A. Cruz

    2006-01-01

    Full Text Available Machine learning is a branch of artificial intelligence that employs a variety of statistical, probabilistic and optimization techniques that allows computers to “learn” from past examples and to detect hard-to-discern patterns from large, noisy or complex data sets. This capability is particularly well-suited to medical applications, especially those that depend on complex proteomic and genomic measurements. As a result, machine learning is frequently used in cancer diagnosis and detection. More recently machine learning has been applied to cancer prognosis and prediction. This latter approach is particularly interesting as it is part of a growing trend towards personalized, predictive medicine. In assembling this review we conducted a broad survey of the different types of machine learning methods being used, the types of data being integrated and the performance of these methods in cancer prediction and prognosis. A number of trends are noted, including a growing dependence on protein biomarkers and microarray data, a strong bias towards applications in prostate and breast cancer, and a heavy reliance on “older” technologies such artificial neural networks (ANNs instead of more recently developed or more easily interpretable machine learning methods. A number of published studies also appear to lack an appropriate level of validation or testing. Among the better designed and validated studies it is clear that machine learning methods can be used to substantially (15-25% improve the accuracy of predicting cancer susceptibility, recurrence and mortality. At a more fundamental level, it is also evident that machine learning is also helping to improve our basic understanding of cancer development and progression.

  15. Differential Prediction Generalization in College Admissions Testing

    Science.gov (United States)

    Aguinis, Herman; Culpepper, Steven A.; Pierce, Charles A.

    2016-01-01

    We introduce the concept of "differential prediction generalization" in the context of college admissions testing. Specifically, we assess the extent to which predicted first-year college grade point average (GPA) based on high-school grade point average (HSGPA) and SAT scores depends on a student's ethnicity and gender and whether this…

  16. Robust block bootstrap panel predictability tests

    NARCIS (Netherlands)

    Westerlund, J.; Smeekes, S.

    2013-01-01

    Most panel data studies of the predictability of returns presume that the cross-sectional units are independent, an assumption that is not realistic. As a response to this, the current paper develops block bootstrap-based panel predictability tests that are valid under very general conditions. Some

  17. Informal Workplace Learning among Nurses: Organisational Learning Conditions and Personal Characteristics That Predict Learning Outcomes

    Science.gov (United States)

    Kyndt, Eva; Vermeire, Eva; Cabus, Shana

    2016-01-01

    Purpose: This paper aims to examine which organisational learning conditions and individual characteristics predict the learning outcomes nurses achieve through informal learning activities. There is specific relevance for the nursing profession because of the rapidly changing healthcare systems. Design/Methodology/Approach: In total, 203 nurses…

  18. Quicksilver: Fast predictive image registration - A deep learning approach.

    Science.gov (United States)

    Yang, Xiao; Kwitt, Roland; Styner, Martin; Niethammer, Marc

    2017-09-01

    This paper introduces Quicksilver, a fast deformable image registration method. Quicksilver registration for image-pairs works by patch-wise prediction of a deformation model based directly on image appearance. A deep encoder-decoder network is used as the prediction model. While the prediction strategy is general, we focus on predictions for the Large Deformation Diffeomorphic Metric Mapping (LDDMM) model. Specifically, we predict the momentum-parameterization of LDDMM, which facilitates a patch-wise prediction strategy while maintaining the theoretical properties of LDDMM, such as guaranteed diffeomorphic mappings for sufficiently strong regularization. We also provide a probabilistic version of our prediction network which can be sampled during the testing time to calculate uncertainties in the predicted deformations. Finally, we introduce a new correction network which greatly increases the prediction accuracy of an already existing prediction network. We show experimental results for uni-modal atlas-to-image as well as uni-/multi-modal image-to-image registrations. These experiments demonstrate that our method accurately predicts registrations obtained by numerical optimization, is very fast, achieves state-of-the-art registration results on four standard validation datasets, and can jointly learn an image similarity measure. Quicksilver is freely available as an open-source software. Copyright © 2017 Elsevier Inc. All rights reserved.

  19. Learning receptive fields using predictive feedback.

    Science.gov (United States)

    Jehee, Janneke F M; Rothkopf, Constantin; Beck, Jeffrey M; Ballard, Dana H

    2006-01-01

    Previously, it was suggested that feedback connections from higher- to lower-level areas carry predictions of lower-level neural activities, whereas feedforward connections carry the residual error between the predictions and the actual lower-level activities [Rao, R.P.N., Ballard, D.H., 1999. Nature Neuroscience 2, 79-87.]. A computational model implementing the hypothesis learned simple cell receptive fields when exposed to natural images. Here, we use predictive feedback to explain tuning properties in medial superior temporal area (MST). We implement the hypothesis using a new, biologically plausible, algorithm based on matching pursuit, which retains all the features of the previous implementation, including its ability to efficiently encode input. When presented with natural images, the model developed receptive field properties as found in primary visual cortex. In addition, when exposed to visual motion input resulting from movements through space, the model learned receptive field properties resembling those in MST. These results corroborate the idea that predictive feedback is a general principle used by the visual system to efficiently encode natural input.

  20. Computerized adaptive testing in computer assisted learning?

    NARCIS (Netherlands)

    Veldkamp, Bernard P.; Matteucci, Mariagiulia; Eggen, Theodorus Johannes Hendrikus Maria; De Wannemacker, Stefan; Clarebout, Geraldine; De Causmaecker, Patrick

    2011-01-01

    A major goal in computerized learning systems is to optimize learning, while in computerized adaptive tests (CAT) efficient measurement of the proficiency of students is the main focus. There seems to be a common interest to integrate computerized adaptive item selection in learning systems and

  1. Prediction of skin sensitization potency using machine learning approaches.

    Science.gov (United States)

    Zang, Qingda; Paris, Michael; Lehmann, David M; Bell, Shannon; Kleinstreuer, Nicole; Allen, David; Matheson, Joanna; Jacobs, Abigail; Casey, Warren; Strickland, Judy

    2017-07-01

    The replacement of animal use in testing for regulatory classification of skin sensitizers is a priority for US federal agencies that use data from such testing. Machine learning models that classify substances as sensitizers or non-sensitizers without using animal data have been developed and evaluated. Because some regulatory agencies require that sensitizers be further classified into potency categories, we developed statistical models to predict skin sensitization potency for murine local lymph node assay (LLNA) and human outcomes. Input variables for our models included six physicochemical properties and data from three non-animal test methods: direct peptide reactivity assay; human cell line activation test; and KeratinoSens™ assay. Models were built to predict three potency categories using four machine learning approaches and were validated using external test sets and leave-one-out cross-validation. A one-tiered strategy modeled all three categories of response together while a two-tiered strategy modeled sensitizer/non-sensitizer responses and then classified the sensitizers as strong or weak sensitizers. The two-tiered model using the support vector machine with all assay and physicochemical data inputs provided the best performance, yielding accuracy of 88% for prediction of LLNA outcomes (120 substances) and 81% for prediction of human test outcomes (87 substances). The best one-tiered model predicted LLNA outcomes with 78% accuracy and human outcomes with 75% accuracy. By comparison, the LLNA predicts human potency categories with 69% accuracy (60 of 87 substances correctly categorized). These results suggest that computational models using non-animal methods may provide valuable information for assessing skin sensitization potency. Copyright © 2017 John Wiley & Sons, Ltd. Copyright © 2017 John Wiley & Sons, Ltd.

  2. PREDICTING ACADEMIC ACHIEVEMENT: THE ROLE OF MOTIVATION AND LEARNING STRATEGIES

    Directory of Open Access Journals (Sweden)

    Juliana Beatriz Stover

    2014-04-01

    Full Text Available The aim of this study consists in testing a predictive model of academic achievement including motivation and learning strategies as predictors. Motivation is defined as the energy and the direction of behaviors; it is categorized in three types of motivation –intrinsic, extrinsic and amotivation (Deci & Ryan, 1985. Learning strategies are deliberate operations oriented towards information processing in academic activities (Valle, Barca, González & Núñez, 1999. Several studies analysed the relationship between motivation and learning strategies in high school and college environments. Students with higher academic achievement were intrinsically motivated and used a wider variety of learning strategies more frequently. A non-experimental predictive design was developed. The sample was composed by 459 students (55.2% high-schoolers; 44.8% college students. Data were gathered by means of sociodemographic and academic surveys, and also by the local versions of the Academic Motivation Scale –EMA, Echelle de Motivation en Éducation (Stover, de la Iglesia, Rial Boubeta & Fernández Liporace, 2012; Vallerand, Blais, Briere & Pelletier, 1989 and the Learning and Study Strategies Inventory –LASSI (Stover, Uriel & Fernández Liporace, 2012; Weinstein, Schulte & Palmer, 1987. Several path analyses were carried out to test a hypothetical model to predict academic achievement (Kline, 1998. Results indicated that self-determined motivation explained academic achievement through the use of learning strategies. The final model obtained an excellent fit (χ2=16.523, df= 6, p=0.011; GFI=0.987; AGFI=0.955; SRMR=0.0320; NFI=0.913; IFI=0.943; CFI=0.940. Results are discussed considering Self Determination Theory and previous research.

  3. Physiognomy: Personality Traits Prediction by Learning

    Institute of Scientific and Technical Information of China (English)

    Ting Zhang; Ri-Zhen Qin; Qiu-Lei Dong; Wei Gao; Hua-Rong Xu; Zhan-Yi Hu

    2017-01-01

    Evaluating individuals' personality traits and intelligence from their faces plays a crucial role in interpersonal relationship and important social events such as elections and court sentences.To assess the possible correlations between personality traits (also measured intelligence) and face images,we first construct a dataset consisting of face photographs,personality measurements,and intelligence measurements.Then,we build an end-to-end convolutional neural network for prediction of personality traits and intelligence to investigate whether self-reported personality traits and intelligence can be predicted reliably from a face image.To our knowledge,it is the first work where deep learning is applied to this problem.Experimental results show the following three points:1)"Rule-consciousness" and "Tension" can be reliably predicted from face images.2) It is difficult,if not impossible,to predict intelligence from face images,a finding in accord with previous studies.3) Convolutional neural network (CNN) features outperform traditional handcrafted features in predicting traits.

  4. Learned Predictiveness Influences Rapid Attentional Capture: Evidence from the Dot Probe Task

    Science.gov (United States)

    Le Pelley, Mike E.; Vadillo, Miguel; Luque, David

    2013-01-01

    Attentional theories of associative learning and categorization propose that learning about the predictiveness of a stimulus influences the amount of attention that is paid to that stimulus. Three experiments tested this idea by looking at the extent to which stimuli that had previously been experienced as predictive or nonpredictive in a…

  5. Predicting Academic Success and Technological Literacy in Secondary Education: A Learning Styles Perspective

    Science.gov (United States)

    Avsec, Stanislav; Szewczyk-Zakrzewska, Agnieszka

    2017-01-01

    This paper aims to investigate the predictive validity of learning styles on academic achievement and technological literacy (TL). For this purpose, secondary school students were recruited (n = 150). An empirical research design was followed where the TL test was used with a learning style inventory measuring learning orientation, processing…

  6. Prediction of beta-turns with learning machines.

    Science.gov (United States)

    Cai, Yu-Dong; Liu, Xiao-Jun; Li, Yi-Xue; Xu, Xue-biao; Chou, Kuo-Chen

    2003-05-01

    The support vector machine approach was introduced to predict the beta-turns in proteins. The overall self-consistency rate by the re-substitution test for the training or learning dataset reached 100%. Both the training dataset and independent testing dataset were taken from Chou [J. Pept. Res. 49 (1997) 120]. The success prediction rates by the jackknife test for the beta-turn subset of 455 tetrapeptides and non-beta-turn subset of 3807 tetrapeptides in the training dataset were 58.1 and 98.4%, respectively. The success rates with the independent dataset test for the beta-turn subset of 110 tetrapeptides and non-beta-turn subset of 30,231 tetrapeptides were 69.1 and 97.3%, respectively. The results obtained from this study support the conclusion that the residue-coupled effect along a tetrapeptide is important for the formation of a beta-turn.

  7. Using Machine Learning to Predict MCNP Bias

    Energy Technology Data Exchange (ETDEWEB)

    Grechanuk, Pavel Aleksandrovi [Los Alamos National Lab. (LANL), Los Alamos, NM (United States)

    2018-01-09

    For many real-world applications in radiation transport where simulations are compared to experimental measurements, like in nuclear criticality safety, the bias (simulated - experimental keff) in the calculation is an extremely important quantity used for code validation. The objective of this project is to accurately predict the bias of MCNP6 [1] criticality calculations using machine learning (ML) algorithms, with the intention of creating a tool that can complement the current nuclear criticality safety methods. In the latest release of MCNP6, the Whisper tool is available for criticality safety analysts and includes a large catalogue of experimental benchmarks, sensitivity profiles, and nuclear data covariance matrices. This data, coming from 1100+ benchmark cases, is used in this study of ML algorithms for criticality safety bias predictions.

  8. Predicting genome-wide redundancy using machine learning

    Directory of Open Access Journals (Sweden)

    Shasha Dennis E

    2010-11-01

    Full Text Available Abstract Background Gene duplication can lead to genetic redundancy, which masks the function of mutated genes in genetic analyses. Methods to increase sensitivity in identifying genetic redundancy can improve the efficiency of reverse genetics and lend insights into the evolutionary outcomes of gene duplication. Machine learning techniques are well suited to classifying gene family members into redundant and non-redundant gene pairs in model species where sufficient genetic and genomic data is available, such as Arabidopsis thaliana, the test case used here. Results Machine learning techniques that combine multiple attributes led to a dramatic improvement in predicting genetic redundancy over single trait classifiers alone, such as BLAST E-values or expression correlation. In withholding analysis, one of the methods used here, Support Vector Machines, was two-fold more precise than single attribute classifiers, reaching a level where the majority of redundant calls were correctly labeled. Using this higher confidence in identifying redundancy, machine learning predicts that about half of all genes in Arabidopsis showed the signature of predicted redundancy with at least one but typically less than three other family members. Interestingly, a large proportion of predicted redundant gene pairs were relatively old duplications (e.g., Ks > 1, suggesting that redundancy is stable over long evolutionary periods. Conclusions Machine learning predicts that most genes will have a functionally redundant paralog but will exhibit redundancy with relatively few genes within a family. The predictions and gene pair attributes for Arabidopsis provide a new resource for research in genetics and genome evolution. These techniques can now be applied to other organisms.

  9. Machine learning modelling for predicting soil liquefaction susceptibility

    Directory of Open Access Journals (Sweden)

    P. Samui

    2011-01-01

    Full Text Available This study describes two machine learning techniques applied to predict liquefaction susceptibility of soil based on the standard penetration test (SPT data from the 1999 Chi-Chi, Taiwan earthquake. The first machine learning technique which uses Artificial Neural Network (ANN based on multi-layer perceptions (MLP that are trained with Levenberg-Marquardt backpropagation algorithm. The second machine learning technique uses the Support Vector machine (SVM that is firmly based on the theory of statistical learning theory, uses classification technique. ANN and SVM have been developed to predict liquefaction susceptibility using corrected SPT [(N160] and cyclic stress ratio (CSR. Further, an attempt has been made to simplify the models, requiring only the two parameters [(N160 and peck ground acceleration (amax/g], for the prediction of liquefaction susceptibility. The developed ANN and SVM models have also been applied to different case histories available globally. The paper also highlights the capability of the SVM over the ANN models.

  10. Intra-and-Inter Species Biomass Prediction in a Plantation Forest: Testing the Utility of High Spatial Resolution Spaceborne Multispectral RapidEye Sensor and Advanced Machine Learning Algorithms

    Directory of Open Access Journals (Sweden)

    Timothy Dube

    2014-08-01

    Full Text Available The quantification of aboveground biomass using remote sensing is critical for better understanding the role of forests in carbon sequestration and for informed sustainable management. Although remote sensing techniques have been proven useful in assessing forest biomass in general, more is required to investigate their capabilities in predicting intra-and-inter species biomass which are mainly characterised by non-linear relationships. In this study, we tested two machine learning algorithms, Stochastic Gradient Boosting (SGB and Random Forest (RF regression trees to predict intra-and-inter species biomass using high resolution RapidEye reflectance bands as well as the derived vegetation indices in a commercial plantation. The results showed that the SGB algorithm yielded the best performance for intra-and-inter species biomass prediction; using all the predictor variables as well as based on the most important selected variables. For example using the most important variables the algorithm produced an R2 of 0.80 and RMSE of 16.93 t·ha−1 for E. grandis; R2 of 0.79, RMSE of 17.27 t·ha−1 for P. taeda and R2 of 0.61, RMSE of 43.39 t·ha−1 for the combined species data sets. Comparatively, RF yielded plausible results only for E. dunii (R2 of 0.79; RMSE of 7.18 t·ha−1. We demonstrated that although the two statistical methods were able to predict biomass accurately, RF produced weaker results as compared to SGB when applied to combined species dataset. The result underscores the relevance of stochastic models in predicting biomass drawn from different species and genera using the new generation high resolution RapidEye sensor with strategically positioned bands.

  11. Machine learning methods in predicting the student academic motivation

    Directory of Open Access Journals (Sweden)

    Ivana Đurđević Babić

    2017-01-01

    Full Text Available Academic motivation is closely related to academic performance. For educators, it is equally important to detect early students with a lack of academic motivation as it is to detect those with a high level of academic motivation. In endeavouring to develop a classification model for predicting student academic motivation based on their behaviour in learning management system (LMS courses, this paper intends to establish links between the predicted student academic motivation and their behaviour in the LMS course. Students from all years at the Faculty of Education in Osijek participated in this research. Three machine learning classifiers (neural networks, decision trees, and support vector machines were used. To establish whether a significant difference in the performance of models exists, a t-test of the difference in proportions was used. Although, all classifiers were successful, the neural network model was shown to be the most successful in detecting the student academic motivation based on their behaviour in LMS course.

  12. Predictive Variable Gain Iterative Learning Control for PMSM

    Directory of Open Access Journals (Sweden)

    Huimin Xu

    2015-01-01

    Full Text Available A predictive variable gain strategy in iterative learning control (ILC is introduced. Predictive variable gain iterative learning control is constructed to improve the performance of trajectory tracking. A scheme based on predictive variable gain iterative learning control for eliminating undesirable vibrations of PMSM system is proposed. The basic idea is that undesirable vibrations of PMSM system are eliminated from two aspects of iterative domain and time domain. The predictive method is utilized to determine the learning gain in the ILC algorithm. Compression mapping principle is used to prove the convergence of the algorithm. Simulation results demonstrate that the predictive variable gain is superior to constant gain and other variable gains.

  13. How to test for partially predictable chaos.

    Science.gov (United States)

    Wernecke, Hendrik; Sándor, Bulcsú; Gros, Claudius

    2017-04-24

    For a chaotic system pairs of initially close-by trajectories become eventually fully uncorrelated on the attracting set. This process of decorrelation can split into an initial exponential decrease and a subsequent diffusive process on the chaotic attractor causing the final loss of predictability. Both processes can be either of the same or of very different time scales. In the latter case the two trajectories linger within a finite but small distance (with respect to the overall extent of the attractor) for exceedingly long times and remain partially predictable. Standard tests for chaos widely use inter-orbital correlations as an indicator. However, testing partially predictable chaos yields mostly ambiguous results, as this type of chaos is characterized by attractors of fractally broadened braids. For a resolution we introduce a novel 0-1 indicator for chaos based on the cross-distance scaling of pairs of initially close trajectories. This test robustly discriminates chaos, including partially predictable chaos, from laminar flow. Additionally using the finite time cross-correlation of pairs of initially close trajectories, we are able to identify laminar flow as well as strong and partially predictable chaos in a 0-1 manner solely from the properties of pairs of trajectories.

  14. Learning and Retention through Predictive Inference and Classification

    Science.gov (United States)

    Sakamoto, Yasuaki; Love, Bradley C.

    2010-01-01

    Work in category learning addresses how humans acquire knowledge and, thus, should inform classroom practices. In two experiments, we apply and evaluate intuitions garnered from laboratory-based research in category learning to learning tasks situated in an educational context. In Experiment 1, learning through predictive inference and…

  15. Learning selenium testing tools with Python

    CERN Document Server

    Gundecha, Unmesh

    2014-01-01

    If you are a quality testing professional, or a software or web application developer looking to create automation test scripts for your web applications, with an interest in Python, then this is the perfect guide for you. Python developers who need to do Selenium testing need not learn Java, as they can directly use Selenium for testing with this book.

  16. Fuzzy-logic based learning style prediction in e-learning using web ...

    Indian Academy of Sciences (India)

    tion, especially in web environments and proposes to use Fuzzy rules to handle the uncertainty in .... learning in safe and supportive environment ... working of the proposed Fuzzy-logic based learning style prediction in e-learning. Section 4.

  17. Prediction of Glucose Tolerance without an Oral Glucose Tolerance Test

    Directory of Open Access Journals (Sweden)

    Rohit Babbar

    2018-03-01

    Full Text Available IntroductionImpaired glucose tolerance (IGT is diagnosed by a standardized oral glucose tolerance test (OGTT. However, the OGTT is laborious, and when not performed, glucose tolerance cannot be determined from fasting samples retrospectively. We tested if glucose tolerance status is reasonably predictable from a combination of demographic, anthropometric, and laboratory data assessed at one time point in a fasting state.MethodsGiven a set of 22 variables selected upon clinical feasibility such as sex, age, height, weight, waist circumference, blood pressure, fasting glucose, HbA1c, hemoglobin, mean corpuscular volume, serum potassium, fasting levels of insulin, C-peptide, triglyceride, non-esterified fatty acids (NEFA, proinsulin, prolactin, cholesterol, low-density lipoprotein, HDL, uric acid, liver transaminases, and ferritin, we used supervised machine learning to estimate glucose tolerance status in 2,337 participants of the TUEF study who were recruited before 2012. We tested the performance of 10 different machine learning classifiers on data from 929 participants in the test set who were recruited after 2012. In addition, reproducibility of IGT was analyzed in 78 participants who had 2 repeated OGTTs within 1 year.ResultsThe most accurate prediction of IGT was reached with the recursive partitioning method (accuracy = 0.78. For all classifiers, mean accuracy was 0.73 ± 0.04. The most important model variable was fasting glucose in all models. Using mean variable importance across all models, fasting glucose was followed by NEFA, triglycerides, HbA1c, and C-peptide. The accuracy of predicting IGT from a previous OGTT was 0.77.ConclusionMachine learning methods yield moderate accuracy in predicting glucose tolerance from a wide set of clinical and laboratory variables. A substitution of OGTT does not currently seem to be feasible. An important constraint could be the limited reproducibility of glucose tolerance status during a

  18. Prediction of Student Dropout in E-Learning Program Through the Use of Machine Learning Method

    Directory of Open Access Journals (Sweden)

    Mingjie Tan

    2015-02-01

    Full Text Available The high rate of dropout is a serious problem in E-learning program. Thus it has received extensive concern from the education administrators and researchers. Predicting the potential dropout students is a workable solution to prevent dropout. Based on the analysis of related literature, this study selected student’s personal characteristic and academic performance as input attributions. Prediction models were developed using Artificial Neural Network (ANN, Decision Tree (DT and Bayesian Networks (BNs. A large sample of 62375 students was utilized in the procedures of model training and testing. The results of each model were presented in confusion matrix, and analyzed by calculating the rates of accuracy, precision, recall, and F-measure. The results suggested all of the three machine learning methods were effective in student dropout prediction, and DT presented a better performance. Finally, some suggestions were made for considerable future research.

  19. Uncovering the neural mechanisms underlying learning from tests.

    Directory of Open Access Journals (Sweden)

    Xiaonan L Liu

    Full Text Available People learn better when re-study opportunities are replaced with tests. While researchers have begun to speculate on why testing is superior to study, few studies have directly examined the neural underpinnings of this effect. In this fMRI study, participants engaged in a study phase to learn arbitrary word pairs, followed by a cued recall test (recall second half of pair when cued with first word of pair, re-study of each pair, and finally another cycle of cued recall tests. Brain activation patterns during the first test (recall of the studied pairs predicts performance on the second test. Importantly, while subsequent memory analyses of encoding trials also predict later accuracy, the brain regions involved in predicting later memory success are more extensive for activity during retrieval (testing than during encoding (study. Those additional regions that predict subsequent memory based on their activation at test but not at encoding may be key to understanding the basis of the testing effect.

  20. Learning and Prediction of Slip from Visual Information

    Science.gov (United States)

    Angelova, Anelia; Matthies, Larry; Helmick, Daniel; Perona, Pietro

    2007-01-01

    This paper presents an approach for slip prediction from a distance for wheeled ground robots using visual information as input. Large amounts of slippage which can occur on certain surfaces, such as sandy slopes, will negatively affect rover mobility. Therefore, obtaining information about slip before entering such terrain can be very useful for better planning and avoiding these areas. To address this problem, terrain appearance and geometry information about map cells are correlated to the slip measured by the rover while traversing each cell. This relationship is learned from previous experience, so slip can be predicted remotely from visual information only. The proposed method consists of terrain type recognition and nonlinear regression modeling. The method has been implemented and tested offline on several off-road terrains including: soil, sand, gravel, and woodchips. The final slip prediction error is about 20%. The system is intended for improved navigation on steep slopes and rough terrain for Mars rovers.

  1. Testing Methodology in the Student Learning Process

    Science.gov (United States)

    Gorbunova, Tatiana N.

    2017-01-01

    The subject of the research is to build methodologies to evaluate the student knowledge by testing. The author points to the importance of feedback about the mastering level in the learning process. Testing is considered as a tool. The object of the study is to create the test system models for defence practice problems. Special attention is paid…

  2. Exploration of Machine Learning Approaches to Predict Pavement Performance

    Science.gov (United States)

    2018-03-23

    Machine learning (ML) techniques were used to model and predict pavement condition index (PCI) for various pavement types using a variety of input variables. The primary objective of this research was to develop and assess PCI predictive models for t...

  3. Learning Digital Test and Diagnostics via Internet

    Directory of Open Access Journals (Sweden)

    Heinz-Dietrich Wuttke

    2007-02-01

    Full Text Available An environment targeted to e-learning is presented for teaching design and test of electronic systems. The environment consists of a set of Java applets, and of web based access to the hardware equipments, which can be used in the classroom, for learning at home, in laboratory research and training, or for carrying out testing of students during exams. The tools support university courses on digital electronics, computer hardware, testing and design for testability to learn by hands-on exercises how to design digital systems, how to make them testable, how to build self-testing systems, how to generate test patterns, how to analyze the quality of tests, and how to localize faults in hardware. The tasks chosen for hands-on training represent simultaneously research problems, which allow to fostering in students critical thinking, problem solving skills and creativity.

  4. An investigation of classification algorithms for predicting HIV drug resistance without genotype resistance testing

    CSIR Research Space (South Africa)

    Brandt, P

    2014-01-01

    Full Text Available is limited in low-resource settings. In this paper we investigate machine learning techniques for drug resistance prediction from routine treatment and laboratory data to help clinicians select patients for confirmatory genotype testing. The techniques...

  5. Springback Prediction on Slit-Ring Test

    International Nuclear Information System (INIS)

    Chen Xiaoming; Shi, Ming F.; Ren Feng; Xia, Z. Cedric

    2005-01-01

    Advanced high strength steels (AHSS) are increasingly being used in the automotive industry to reduce vehicle weight while improving vehicle crash performance. One of the concerns in manufacturing is springback control after stamping. Although computer simulation technologies have been successfully applied to predict stamping formability, they still face major challenges in springback prediction, particularly for AHSS. Springback analysis is very complicated and involves large deformation problems in the forming stage and mechanical multiplying effect during the elastic recovery after releasing a part from the die. Therefore, the predictions are very sensitive to the simulation parameters used. It is very critical in springback simulation to choose an appropriate material model, element formulation and contact algorithm. In this study, a springback benchmark test, the slit ring cup, is used in the springback simulation with commercially available finite element analysis (FEA) software, LS-DYNA. The sensitivity of seven simulation variables on springback predictions was investigated, and a set of parameters with stable simulation results was identified. Final simulations using the selected set of parameters were conducted on six different materials including two AHSS steels, two conventional high strength steels, one mild steel and an aluminum alloy. The simulation results are compared with experimental measurements for all six materials and a favorable result is achieved. Simulation errors as compared against test results falls within 10%

  6. Machine learning landscapes and predictions for patient outcomes

    Science.gov (United States)

    Das, Ritankar; Wales, David J.

    2017-07-01

    The theory and computational tools developed to interpret and explore energy landscapes in molecular science are applied to the landscapes defined by local minima for neural networks. These machine learning landscapes correspond to fits of training data, where the inputs are vital signs and laboratory measurements for a database of patients, and the objective is to predict a clinical outcome. In this contribution, we test the predictions obtained by fitting to single measurements, and then to combinations of between 2 and 10 different patient medical data items. The effect of including measurements over different time intervals from the 48 h period in question is analysed, and the most recent values are found to be the most important. We also compare results obtained for neural networks as a function of the number of hidden nodes, and for different values of a regularization parameter. The predictions are compared with an alternative convex fitting function, and a strong correlation is observed. The dependence of these results on the patients randomly selected for training and testing decreases systematically with the size of the database available. The machine learning landscapes defined by neural network fits in this investigation have single-funnel character, which probably explains why it is relatively straightforward to obtain the global minimum solution, or a fit that behaves similarly to this optimal parameterization.

  7. Gene Prediction in Metagenomic Fragments with Deep Learning

    Directory of Open Access Journals (Sweden)

    Shao-Wu Zhang

    2017-01-01

    Full Text Available Next generation sequencing technologies used in metagenomics yield numerous sequencing fragments which come from thousands of different species. Accurately identifying genes from metagenomics fragments is one of the most fundamental issues in metagenomics. In this article, by fusing multifeatures (i.e., monocodon usage, monoamino acid usage, ORF length coverage, and Z-curve features and using deep stacking networks learning model, we present a novel method (called Meta-MFDL to predict the metagenomic genes. The results with 10 CV and independent tests show that Meta-MFDL is a powerful tool for identifying genes from metagenomic fragments.

  8. Stress before extinction learning enhances and generalizes extinction memory in a predictive learning task.

    Science.gov (United States)

    Meir Drexler, Shira; Hamacher-Dang, Tanja C; Wolf, Oliver T

    2017-05-01

    In extinction learning, the individual learns that a previously acquired association (e.g. between a threat and its predictor) is no longer valid. This learning is the principle underlying many cognitive-behavioral psychotherapeutic treatments, e.g. 'exposure therapy'. However, extinction is often highly-context dependent, leading to renewal (relapse of extinguished conditioned response following context change). We have previously shown that post-extinction stress leads to a more context-dependent extinction memory in a predictive learning task. Yet as stress prior to learning can impair the integration of contextual cues, here we aim to create a more generalized extinction memory by inducing stress prior to extinction. Forty-nine men and women learned the associations between stimuli and outcomes in a predictive learning task (day 1), extinguished them shortly after an exposure to a stress/control condition (day 2), and were tested for renewal (day 3). No group differences were seen in acquisition and extinction learning, and a renewal effect was present in both groups. However, the groups differed in the strength and context-dependency of the extinction memory. Compared to the control group, the stress group showed an overall reduced recovery of responding to the extinguished stimuli, in particular in the acquisition context. These results, together with our previous findings, demonstrate that the effects of stress exposure on extinction memory depend on its timing. While post-extinction stress makes the memory more context-bound, pre-extinction stress strengthens its consolidation for the acquisition context as well, making it potentially more resistant to relapse. These results have implications for the use of glucocorticoids as extinction-enhancers in exposure therapy. Copyright © 2017 Elsevier Inc. All rights reserved.

  9. Reminder cues modulate the renewal effect in human predictive learning

    Directory of Open Access Journals (Sweden)

    Javier Bustamante

    2016-12-01

    Full Text Available Associative learning refers to our ability to learn about regularities in our environment. When a stimulus is repeatedly followed by a specific outcome, we learn to expect the outcome in the presence of the stimulus. We are also able to modify established expectations in the face of disconfirming information (the stimulus is no longer followed by the outcome. Both the change of environmental regularities and the related processes of adaptation are referred to as extinction. However, extinction does not erase the initially acquired expectations. For instance, following successful extinction, the initially learned expectations can recover when there is a context change – a phenomenon called the renewal effect, which is considered as a model for relapse after exposure therapy. Renewal was found to be modulated by reminder cues of acquisition and extinction. However, the mechanisms underlying the effectiveness of reminder cues are not well understood. The aim of the present study was to investigate the impact of reminder cues on renewal in the field of human predictive learning. Experiment I demonstrated that renewal in human predictive learning is modulated by cues related to acquisition or extinction. Initially, participants received pairings of a stimulus and an outcome in one context. These stimulus-outcome pairings were preceded by presentations of a reminder cue (acquisition cue. Then, participants received extinction in a different context in which presentations of the stimulus were no longer followed by the outcome. These extinction trials were preceded by a second reminder cue (extinction cue. During a final phase conducted in a third context, participants showed stronger expectations of the outcome in the presence of the stimulus when testing was accompanied by the acquisition cue compared to the extinction cue. Experiment II tested an explanation of the reminder cue effect in terms of simple cue-outcome associations. Therefore

  10. Pretest Predictions for Phase II Ventilation Tests

    International Nuclear Information System (INIS)

    Yiming Sun

    2001-01-01

    The objective of this calculation is to predict the temperatures of the ventilating air, waste package surface, and concrete pipe walls that will be developed during the Phase II ventilation tests involving various test conditions. The results will be used as inputs to validating numerical approach for modeling continuous ventilation, and be used to support the repository subsurface design. The scope of the calculation is to identify the physical mechanisms and parameters related to thermal response in the Phase II ventilation tests, and describe numerical methods that are used to calculate the effects of continuous ventilation. The calculation is limited to thermal effect only. This engineering work activity is conducted in accordance with the ''Technical Work Plan for: Subsurface Performance Testing for License Application (LA) for Fiscal Year 2001'' (CRWMS M and O 2000d). This technical work plan (TWP) includes an AP-2.21Q, ''Quality Determinations and Planning for Scientific, Engineering, and Regulatory Compliance Activities'', activity evaluation (CRWMS M and O 2000d, Addendum A) that has determined this activity is subject to the YMP quality assurance (QA) program. The calculation is developed in accordance with the AP-3.12Q procedure, ''Calculations''. Additional background information regarding this activity is contained in the ''Development Plan for Ventilation Pretest Predictive Calculation'' (DP) (CRWMS M and O 2000a)

  11. Implicit Learning Abilities Predict Treatment Response in Autism Spectrum Disorders

    Science.gov (United States)

    2015-09-01

    early behavioral interventions are the most effective treatment for Autism Spectrum Disorder (ASD), but almost half of the children do not make...behavioral intervention . 2. KEYWORDS Autism Spectrum Disorder , implicit learning, associative learning, individual differences, functional Magnetic...2 AWARD NUMBER: W81XWH-14-1-0261 TITLE: Implicit Learning Abilities Predict Treatment Response in Autism Spectrum Disorders PRINCIPAL

  12. Predicting Student Performance in a Collaborative Learning Environment

    Science.gov (United States)

    Olsen, Jennifer K.; Aleven, Vincent; Rummel, Nikol

    2015-01-01

    Student models for adaptive systems may not model collaborative learning optimally. Past research has either focused on modeling individual learning or for collaboration, has focused on group dynamics or group processes without predicting learning. In the current paper, we adjust the Additive Factors Model (AFM), a standard logistic regression…

  13. A novel time series link prediction method: Learning automata approach

    Science.gov (United States)

    Moradabadi, Behnaz; Meybodi, Mohammad Reza

    2017-09-01

    Link prediction is a main social network challenge that uses the network structure to predict future links. The common link prediction approaches to predict hidden links use a static graph representation where a snapshot of the network is analyzed to find hidden or future links. For example, similarity metric based link predictions are a common traditional approach that calculates the similarity metric for each non-connected link and sort the links based on their similarity metrics and label the links with higher similarity scores as the future links. Because people activities in social networks are dynamic and uncertainty, and the structure of the networks changes over time, using deterministic graphs for modeling and analysis of the social network may not be appropriate. In the time-series link prediction problem, the time series link occurrences are used to predict the future links In this paper, we propose a new time series link prediction based on learning automata. In the proposed algorithm for each link that must be predicted there is one learning automaton and each learning automaton tries to predict the existence or non-existence of the corresponding link. To predict the link occurrence in time T, there is a chain consists of stages 1 through T - 1 and the learning automaton passes from these stages to learn the existence or non-existence of the corresponding link. Our preliminary link prediction experiments with co-authorship and email networks have provided satisfactory results when time series link occurrences are considered.

  14. Conformal prediction for reliable machine learning theory, adaptations and applications

    CERN Document Server

    Balasubramanian, Vineeth; Vovk, Vladimir

    2014-01-01

    The conformal predictions framework is a recent development in machine learning that can associate a reliable measure of confidence with a prediction in any real-world pattern recognition application, including risk-sensitive applications such as medical diagnosis, face recognition, and financial risk prediction. Conformal Predictions for Reliable Machine Learning: Theory, Adaptations and Applications captures the basic theory of the framework, demonstrates how to apply it to real-world problems, and presents several adaptations, including active learning, change detection, and anomaly detecti

  15. Predicting DPP-IV inhibitors with machine learning approaches

    Science.gov (United States)

    Cai, Jie; Li, Chanjuan; Liu, Zhihong; Du, Jiewen; Ye, Jiming; Gu, Qiong; Xu, Jun

    2017-04-01

    Dipeptidyl peptidase IV (DPP-IV) is a promising Type 2 diabetes mellitus (T2DM) drug target. DPP-IV inhibitors prolong the action of glucagon-like peptide-1 (GLP-1) and gastric inhibitory peptide (GIP), improve glucose homeostasis without weight gain, edema, and hypoglycemia. However, the marketed DPP-IV inhibitors have adverse effects such as nasopharyngitis, headache, nausea, hypersensitivity, skin reactions and pancreatitis. Therefore, it is still expected for novel DPP-IV inhibitors with minimal adverse effects. The scaffolds of existing DPP-IV inhibitors are structurally diversified. This makes it difficult to build virtual screening models based upon the known DPP-IV inhibitor libraries using conventional QSAR approaches. In this paper, we report a new strategy to predict DPP-IV inhibitors with machine learning approaches involving naïve Bayesian (NB) and recursive partitioning (RP) methods. We built 247 machine learning models based on 1307 known DPP-IV inhibitors with optimized molecular properties and topological fingerprints as descriptors. The overall predictive accuracies of the optimized models were greater than 80%. An external test set, composed of 65 recently reported compounds, was employed to validate the optimized models. The results demonstrated that both NB and RP models have a good predictive ability based on different combinations of descriptors. Twenty "good" and twenty "bad" structural fragments for DPP-IV inhibitors can also be derived from these models for inspiring the new DPP-IV inhibitor scaffold design.

  16. Do functional tests predict low back pain?

    Science.gov (United States)

    Takala, E P; Viikari-Juntura, E

    2000-08-15

    A cohort of 307 nonsymptomatic workers and another cohort of 123 workers with previous episodes of low back pain were followed up for 2 years. The outcomes were measured by symptoms, medical consultations, and sick leaves due to low back disorders. To study the predictive value of a set of tests measuring the physical performance of the back in a working population. The hypothesis was that subjects with poor functional capacity are liable to back disorders. Reduced functional performance has been associated with back pain. There are few data to show whether reduced functional capacity is a cause or a consequence of pain. Mobility of the trunk in forward and side bending, maximal isokinetic trunk extension, flexion and lifting strength, and static endurance of back extension were measured. Standing balance and foot reaction time were recorded with a force plate. Clinical tests for the provocation of back or leg pain were performed. Gender, workload, age, and anthropometrics were managed as potential confounders in the analysis. Marked overlapping was seen in the measures of the subjects with different outcomes. Among the nonsymptomatic subjects, low performance in tests of mobility and standing balance was associated with future back disorders. Among workers with previous episodes of back pain, low isokinetic extension strength, poor standing balance, and positive clinical signs predicted future pain. Some associations were found between the functional tests and future low back pain. The wide variation in the results questions the value of the tests in health examinations (e.g., in screening or surveillance of low back disorders).

  17. Machine-Learning-Based No Show Prediction in Outpatient Visits

    Directory of Open Access Journals (Sweden)

    Carlos Elvira

    2018-03-01

    Full Text Available A recurring problem in healthcare is the high percentage of patients who miss their appointment, be it a consultation or a hospital test. The present study seeks patient’s behavioural patterns that allow predicting the probability of no- shows. We explore the convenience of using Big Data Machine Learning models to accomplish this task. To begin with, a predictive model based only on variables associated with the target appointment is built. Then the model is improved by considering the patient’s history of appointments. In both cases, the Gradient Boosting algorithm was the predictor of choice. Our numerical results are considered promising given the small amount of information available. However, there seems to be plenty of room to improve the model if we manage to collect additional data for both patients and appointments.

  18. Predicting Smoking Status Using Machine Learning Algorithms and Statistical Analysis

    Directory of Open Access Journals (Sweden)

    Charles Frank

    2018-03-01

    Full Text Available Smoking has been proven to negatively affect health in a multitude of ways. As of 2009, smoking has been considered the leading cause of preventable morbidity and mortality in the United States, continuing to plague the country’s overall health. This study aims to investigate the viability and effectiveness of some machine learning algorithms for predicting the smoking status of patients based on their blood tests and vital readings results. The analysis of this study is divided into two parts: In part 1, we use One-way ANOVA analysis with SAS tool to show the statistically significant difference in blood test readings between smokers and non-smokers. The results show that the difference in INR, which measures the effectiveness of anticoagulants, was significant in favor of non-smokers which further confirms the health risks associated with smoking. In part 2, we use five machine learning algorithms: Naïve Bayes, MLP, Logistic regression classifier, J48 and Decision Table to predict the smoking status of patients. To compare the effectiveness of these algorithms we use: Precision, Recall, F-measure and Accuracy measures. The results show that the Logistic algorithm outperformed the four other algorithms with Precision, Recall, F-Measure, and Accuracy of 83%, 83.4%, 83.2%, 83.44%, respectively.

  19. Statistical and Machine Learning Models to Predict Programming Performance

    OpenAIRE

    Bergin, Susan

    2006-01-01

    This thesis details a longitudinal study on factors that influence introductory programming success and on the development of machine learning models to predict incoming student performance. Although numerous studies have developed models to predict programming success, the models struggled to achieve high accuracy in predicting the likely performance of incoming students. Our approach overcomes this by providing a machine learning technique, using a set of three significant...

  20. Prediction of stroke thrombolysis outcome using CT brain machine learning

    Directory of Open Access Journals (Sweden)

    Paul Bentley

    2014-01-01

    Full Text Available A critical decision-step in the emergency treatment of ischemic stroke is whether or not to administer thrombolysis — a treatment that can result in good recovery, or deterioration due to symptomatic intracranial haemorrhage (SICH. Certain imaging features based upon early computerized tomography (CT, in combination with clinical variables, have been found to predict SICH, albeit with modest accuracy. In this proof-of-concept study, we determine whether machine learning of CT images can predict which patients receiving tPA will develop SICH as opposed to showing clinical improvement with no haemorrhage. Clinical records and CT brains of 116 acute ischemic stroke patients treated with intravenous thrombolysis were collected retrospectively (including 16 who developed SICH. The sample was split into training (n = 106 and test sets (n = 10, repeatedly for 1760 different combinations. CT brain images acted as inputs into a support vector machine (SVM, along with clinical severity. Performance of the SVM was compared with established prognostication tools (SEDAN and HAT scores; original, or after adaptation to our cohort. Predictive performance, assessed as area under receiver-operating-characteristic curve (AUC, of the SVM (0.744 compared favourably with that of prognostic scores (original and adapted versions: 0.626–0.720; p < 0.01. The SVM also identified 9 out of 16 SICHs, as opposed to 1–5 using prognostic scores, assuming a 10% SICH frequency (p < 0.001. In summary, machine learning methods applied to acute stroke CT images offer automation, and potentially improved performance, for prediction of SICH following thrombolysis. Larger-scale cohorts, and incorporation of advanced imaging, should be tested with such methods.

  1. Predicting the concentration of residual methanol in industrial formalin using machine learning

    OpenAIRE

    Heidkamp, William

    2016-01-01

    In this thesis, a machine learning approach was used to develop a predictive model for residual methanol concentration in industrial formalin produced at the Akzo Nobel factory in Kristinehamn, Sweden. The MATLABTM computational environment supplemented with the Statistics and Machine LearningTM toolbox from the MathWorks were used to test various machine learning algorithms on the formalin production data from Akzo Nobel. As a result, the Gaussian Process Regression algorithm was found to pr...

  2. Hebbian Learning is about contingency, not contiguity, and explains the emergence of predictive mirror neurons

    NARCIS (Netherlands)

    Keysers, C.; Perrett, David I; Gazzola, Valeria

    Hebbian Learning should not be reduced to contiguity, as it detects contingency and causality. Hebbian Learning accounts of mirror neurons make predictions that differ from associative learning: Through Hebbian Learning, mirror neurons become dynamic networks that calculate predictions and

  3. Deep learning methods for protein torsion angle prediction.

    Science.gov (United States)

    Li, Haiou; Hou, Jie; Adhikari, Badri; Lyu, Qiang; Cheng, Jianlin

    2017-09-18

    Deep learning is one of the most powerful machine learning methods that has achieved the state-of-the-art performance in many domains. Since deep learning was introduced to the field of bioinformatics in 2012, it has achieved success in a number of areas such as protein residue-residue contact prediction, secondary structure prediction, and fold recognition. In this work, we developed deep learning methods to improve the prediction of torsion (dihedral) angles of proteins. We design four different deep learning architectures to predict protein torsion angles. The architectures including deep neural network (DNN) and deep restricted Boltzmann machine (DRBN), deep recurrent neural network (DRNN) and deep recurrent restricted Boltzmann machine (DReRBM) since the protein torsion angle prediction is a sequence related problem. In addition to existing protein features, two new features (predicted residue contact number and the error distribution of torsion angles extracted from sequence fragments) are used as input to each of the four deep learning architectures to predict phi and psi angles of protein backbone. The mean absolute error (MAE) of phi and psi angles predicted by DRNN, DReRBM, DRBM and DNN is about 20-21° and 29-30° on an independent dataset. The MAE of phi angle is comparable to the existing methods, but the MAE of psi angle is 29°, 2° lower than the existing methods. On the latest CASP12 targets, our methods also achieved the performance better than or comparable to a state-of-the art method. Our experiment demonstrates that deep learning is a valuable method for predicting protein torsion angles. The deep recurrent network architecture performs slightly better than deep feed-forward architecture, and the predicted residue contact number and the error distribution of torsion angles extracted from sequence fragments are useful features for improving prediction accuracy.

  4. Predictable Locations Aid Early Object Name Learning

    Science.gov (United States)

    Benitez, Viridiana L.; Smith, Linda B.

    2012-01-01

    Expectancy-based localized attention has been shown to promote the formation and retrieval of multisensory memories in adults. Three experiments show that these processes also characterize attention and learning in 16- to 18-month old infants and, moreover, that these processes may play a critical role in supporting early object name learning. The…

  5. Prostate Cancer Probability Prediction By Machine Learning Technique.

    Science.gov (United States)

    Jović, Srđan; Miljković, Milica; Ivanović, Miljan; Šaranović, Milena; Arsić, Milena

    2017-11-26

    The main goal of the study was to explore possibility of prostate cancer prediction by machine learning techniques. In order to improve the survival probability of the prostate cancer patients it is essential to make suitable prediction models of the prostate cancer. If one make relevant prediction of the prostate cancer it is easy to create suitable treatment based on the prediction results. Machine learning techniques are the most common techniques for the creation of the predictive models. Therefore in this study several machine techniques were applied and compared. The obtained results were analyzed and discussed. It was concluded that the machine learning techniques could be used for the relevant prediction of prostate cancer.

  6. The effect of testing on skills learning

    DEFF Research Database (Denmark)

    Kromann, Charles B; Jensen, Morten L; Ringsted, Charlotte

    2009-01-01

    OBJECTIVES: In addition to the extrinsic effects of assessment and examinations on students' study habits, testing can have an intrinsic effect on the memory of studied material. Whether this testing effect also applies to skills learning is not known. However, this is especially interesting...... a prospective, controlled, randomised, single-blind, post-test-only intervention study, preceded by a similar pre- and post-test pilot study in order to make a power calculation. A total of 140 medical students participating in a mandatory 4-hour in-hospital resuscitation course in the seventh semester were...

  7. Predictive acute toxicity tests with pesticides.

    Science.gov (United States)

    Brown, V K

    1983-01-01

    By definition pesticides are biocidal products and this implies a probability that pesticides may be acutely toxic to species other than the designated target species. The ways in which pesticides are manufactured, formulated, packaged, distributed and used necessitates a potential for the exposure of non-target species although the technology exists to minimize adventitious exposure. The occurrence of deliberate exposure of non-target species due to the misuse of pesticides is known to happen. The array of predictive acute toxicity tests carried out on pesticides and involving the use of laboratory animals can be justified as providing data on which hazard assessment can be based. This paper addresses the justification and rationale of this statement.

  8. Computational neurorehabilitation: modeling plasticity and learning to predict recovery.

    Science.gov (United States)

    Reinkensmeyer, David J; Burdet, Etienne; Casadio, Maura; Krakauer, John W; Kwakkel, Gert; Lang, Catherine E; Swinnen, Stephan P; Ward, Nick S; Schweighofer, Nicolas

    2016-04-30

    Despite progress in using computational approaches to inform medicine and neuroscience in the last 30 years, there have been few attempts to model the mechanisms underlying sensorimotor rehabilitation. We argue that a fundamental understanding of neurologic recovery, and as a result accurate predictions at the individual level, will be facilitated by developing computational models of the salient neural processes, including plasticity and learning systems of the brain, and integrating them into a context specific to rehabilitation. Here, we therefore discuss Computational Neurorehabilitation, a newly emerging field aimed at modeling plasticity and motor learning to understand and improve movement recovery of individuals with neurologic impairment. We first explain how the emergence of robotics and wearable sensors for rehabilitation is providing data that make development and testing of such models increasingly feasible. We then review key aspects of plasticity and motor learning that such models will incorporate. We proceed by discussing how computational neurorehabilitation models relate to the current benchmark in rehabilitation modeling - regression-based, prognostic modeling. We then critically discuss the first computational neurorehabilitation models, which have primarily focused on modeling rehabilitation of the upper extremity after stroke, and show how even simple models have produced novel ideas for future investigation. Finally, we conclude with key directions for future research, anticipating that soon we will see the emergence of mechanistic models of motor recovery that are informed by clinical imaging results and driven by the actual movement content of rehabilitation therapy as well as wearable sensor-based records of daily activity.

  9. Prediction of Baseflow Index of Catchments using Machine Learning Algorithms

    Science.gov (United States)

    Yadav, B.; Hatfield, K.

    2017-12-01

    We present the results of eight machine learning techniques for predicting the baseflow index (BFI) of ungauged basins using a surrogate of catchment scale climate and physiographic data. The tested algorithms include ordinary least squares, ridge regression, least absolute shrinkage and selection operator (lasso), elasticnet, support vector machine, gradient boosted regression trees, random forests, and extremely randomized trees. Our work seeks to identify the dominant controls of BFI that can be readily obtained from ancillary geospatial databases and remote sensing measurements, such that the developed techniques can be extended to ungauged catchments. More than 800 gauged catchments spanning the continental United States were selected to develop the general methodology. The BFI calculation was based on the baseflow separated from daily streamflow hydrograph using HYSEP filter. The surrogate catchment attributes were compiled from multiple sources including digital elevation model, soil, landuse, climate data, other publicly available ancillary and geospatial data. 80% catchments were used to train the ML algorithms, and the remaining 20% of the catchments were used as an independent test set to measure the generalization performance of fitted models. A k-fold cross-validation using exhaustive grid search was used to fit the hyperparameters of each model. Initial model development was based on 19 independent variables, but after variable selection and feature ranking, we generated revised sparse models of BFI prediction that are based on only six catchment attributes. These key predictive variables selected after the careful evaluation of bias-variance tradeoff include average catchment elevation, slope, fraction of sand, permeability, temperature, and precipitation. The most promising algorithms exceeding an accuracy score (r-square) of 0.7 on test data include support vector machine, gradient boosted regression trees, random forests, and extremely randomized

  10. Learned predictiveness and outcome predictability effects are not simply two sides of the same coin.

    Science.gov (United States)

    Thorwart, Anna; Livesey, Evan J; Wilhelm, Francisco; Liu, Wei; Lachnit, Harald

    2017-10-01

    The Learned Predictiveness effect refers to the observation that learning about the relationship between a cue and an outcome is influenced by the predictive relevance of the cue for other outcomes. Similarly, the Outcome Predictability effect refers to a recent observation that the previous predictability of an outcome affects learning about this outcome in new situations, too. We hypothesize that both effects may be two manifestations of the same phenomenon and stimuli that have been involved in highly predictive relationships may be learned about faster when they are involved in new relationships regardless of their functional role in predictive learning as cues and outcomes. Four experiments manipulated both the relationships and the function of the stimuli. While we were able to replicate the standard effects, they did not survive a transfer to situations where the functional role of the stimuli changed, that is the outcome of the first phase becomes a cue in the second learning phase or the cue of the first phase becomes the outcome of the second phase. Furthermore, unlike learned predictiveness, there was little indication that the distribution of overt attention in the second phase was influenced by previous predictability. The results suggest that these 2 very similar effects are not manifestations of a more general phenomenon but rather independent from each other. (PsycINFO Database Record (c) 2017 APA, all rights reserved).

  11. Using machine learning, neural networks and statistics to predict bankruptcy

    NARCIS (Netherlands)

    Pompe, P.P.M.; Feelders, A.J.; Feelders, A.J.

    1997-01-01

    Recent literature strongly suggests that machine learning approaches to classification outperform "classical" statistical methods. We make a comparison between the performance of linear discriminant analysis, classification trees, and neural networks in predicting corporate bankruptcy. Linear

  12. A deep learning-based multi-model ensemble method for cancer prediction.

    Science.gov (United States)

    Xiao, Yawen; Wu, Jun; Lin, Zongli; Zhao, Xiaodong

    2018-01-01

    Cancer is a complex worldwide health problem associated with high mortality. With the rapid development of the high-throughput sequencing technology and the application of various machine learning methods that have emerged in recent years, progress in cancer prediction has been increasingly made based on gene expression, providing insight into effective and accurate treatment decision making. Thus, developing machine learning methods, which can successfully distinguish cancer patients from healthy persons, is of great current interest. However, among the classification methods applied to cancer prediction so far, no one method outperforms all the others. In this paper, we demonstrate a new strategy, which applies deep learning to an ensemble approach that incorporates multiple different machine learning models. We supply informative gene data selected by differential gene expression analysis to five different classification models. Then, a deep learning method is employed to ensemble the outputs of the five classifiers. The proposed deep learning-based multi-model ensemble method was tested on three public RNA-seq data sets of three kinds of cancers, Lung Adenocarcinoma, Stomach Adenocarcinoma and Breast Invasive Carcinoma. The test results indicate that it increases the prediction accuracy of cancer for all the tested RNA-seq data sets as compared to using a single classifier or the majority voting algorithm. By taking full advantage of different classifiers, the proposed deep learning-based multi-model ensemble method is shown to be accurate and effective for cancer prediction. Copyright © 2017 Elsevier B.V. All rights reserved.

  13. Resting alpha activity predicts learning ability in alpha neurofeedback

    Directory of Open Access Journals (Sweden)

    Wenya eNan

    2014-07-01

    Full Text Available Individuals differ in their ability to learn how to regulate the alpha activity by neurofeedback. This study aimed to investigate whether the resting alpha activity is related to the learning ability of alpha enhancement in neurofeedback and could be used as a predictor. A total of 25 subjects performed 20 sessions of individualized alpha neurofeedback in order to learn how to enhance activity in the alpha frequency band. The learning ability was assessed by three indices respectively: the training parameter changes between two periods, within a short period and across the whole training time. It was found that the resting alpha amplitude measured before training had significant positive correlations with all learning indices and could be used as a predictor for the learning ability prediction. This finding would help the researchers in not only predicting the training efficacy in individuals but also gaining further insight into the mechanisms of alpha neurofeedback.

  14. Probabilistic Category Learning in Developmental Dyslexia: Evidence from Feedback and Paired-Associate Weather Prediction Tasks

    Science.gov (United States)

    Gabay, Yafit; Vakil, Eli; Schiff, Rachel; Holt, Lori L.

    2015-01-01

    Objective Developmental dyslexia is presumed to arise from specific phonological impairments. However, an emerging theoretical framework suggests that phonological impairments may be symptoms stemming from an underlying dysfunction of procedural learning. Method We tested procedural learning in adults with dyslexia (n=15) and matched-controls (n=15) using two versions of the Weather Prediction Task: Feedback (FB) and Paired-associate (PA). In the FB-based task, participants learned associations between cues and outcomes initially by guessing and subsequently through feedback indicating the correctness of response. In the PA-based learning task, participants viewed the cue and its associated outcome simultaneously without overt response or feedback. In both versions, participants trained across 150 trials. Learning was assessed in a subsequent test without presentation of the outcome, or corrective feedback. Results The Dyslexia group exhibited impaired learning compared with the Control group on both the FB and PA versions of the weather prediction task. Conclusions The results indicate that the ability to learn by feedback is not selectively impaired in dyslexia. Rather it seems that the probabilistic nature of the task, shared by the FB and PA versions of the weather prediction task, hampers learning in those with dyslexia. Results are discussed in light of procedural learning impairments among participants with dyslexia. PMID:25730732

  15. Machine learning applied to the prediction of citrus production

    OpenAIRE

    Díaz Rodríguez, Susana Irene; Mazza, Silvia M.; Fernández-Combarro Álvarez, Elías; Giménez, Laura I.; Gaiad, José E.

    2017-01-01

    An in-depth knowledge about variables affecting production is required in order to predict global production and take decisions in agriculture. Machine learning is a technique used in agricultural planning and precision agriculture. This work (i) studies the effectiveness of machine learning techniques for predicting orchards production; and (ii) variables affecting this production were also identified. Data from 964 orchards of lemon, mandarin, and orange in Corrientes, Argentina are analyse...

  16. Machine learning applied to crime prediction

    OpenAIRE

    Vaquero Barnadas, Miquel

    2016-01-01

    Machine Learning is a cornerstone when it comes to artificial intelligence and big data analysis. It provides powerful algorithms that are capable of recognizing patterns, classifying data, and, basically, learn by themselves to perform a specific task. This field has incredibly grown in popularity these days, however, it still remains unknown for the majority of people, and even for most professionals. This project intends to provide an understandable explanation of what is it, what types ar...

  17. Learning Behavior Models for Interpreting and Predicting Traffic Situations

    OpenAIRE

    Gindele, Tobias

    2014-01-01

    In this thesis, we present Bayesian state estimation and machine learning methods for predicting traffic situations. The cognitive ability to assess situations and behaviors of traffic participants, and to anticipate possible developments is an essential requirement for several applications in the traffic domain, especially for self-driving cars. We present a method for learning behavior models from unlabeled traffic observations and develop improved learning methods for decision trees.

  18. Automatically explaining machine learning prediction results: a demonstration on type 2 diabetes risk prediction.

    Science.gov (United States)

    Luo, Gang

    2016-01-01

    Predictive modeling is a key component of solutions to many healthcare problems. Among all predictive modeling approaches, machine learning methods often achieve the highest prediction accuracy, but suffer from a long-standing open problem precluding their widespread use in healthcare. Most machine learning models give no explanation for their prediction results, whereas interpretability is essential for a predictive model to be adopted in typical healthcare settings. This paper presents the first complete method for automatically explaining results for any machine learning predictive model without degrading accuracy. We did a computer coding implementation of the method. Using the electronic medical record data set from the Practice Fusion diabetes classification competition containing patient records from all 50 states in the United States, we demonstrated the method on predicting type 2 diabetes diagnosis within the next year. For the champion machine learning model of the competition, our method explained prediction results for 87.4 % of patients who were correctly predicted by the model to have type 2 diabetes diagnosis within the next year. Our demonstration showed the feasibility of automatically explaining results for any machine learning predictive model without degrading accuracy.

  19. On-Line, Self-Learning, Predictive Tool for Determining Payload Thermal Response

    Science.gov (United States)

    Jen, Chian-Li; Tilwick, Leon

    2000-01-01

    This paper will present the results of a joint ManTech / Goddard R&D effort, currently under way, to develop and test a computer based, on-line, predictive simulation model for use by facility operators to predict the thermal response of a payload during thermal vacuum testing. Thermal response was identified as an area that could benefit from the algorithms developed by Dr. Jeri for complex computer simulations. Most thermal vacuum test setups are unique since no two payloads have the same thermal properties. This requires that the operators depend on their past experiences to conduct the test which requires time for them to learn how the payload responds while at the same time limiting any risk of exceeding hot or cold temperature limits. The predictive tool being developed is intended to be used with the new Thermal Vacuum Data System (TVDS) developed at Goddard for the Thermal Vacuum Test Operations group. This model can learn the thermal response of the payload by reading a few data points from the TVDS, accepting the payload's current temperature as the initial condition for prediction. The model can then be used as a predictive tool to estimate the future payload temperatures according to a predetermined shroud temperature profile. If the error of prediction is too big, the model can be asked to re-learn the new situation on-line in real-time and give a new prediction. Based on some preliminary tests, we feel this predictive model can forecast the payload temperature of the entire test cycle within 5 degrees Celsius after it has learned 3 times during the beginning of the test. The tool will allow the operator to play "what-if' experiments to decide what is his best shroud temperature set-point control strategy. This tool will save money by minimizing guess work and optimizing transitions as well as making the testing process safer and easier to conduct.

  20. Predictive radiosensitivity tests in human lymphocytes

    International Nuclear Information System (INIS)

    Di Giorgio, Marina; Vallerga, Maria B.; Taja, Maria R.; Sardi, M.; Busto, E.; Mairal, L.; Roth, B.; Menendez, P.; Bonomi, M.

    2004-01-01

    Individual radiosensitivity is an inherent characteristic, associated with an abnormally increased reaction to ionising radiation of both the whole body and cells derived from body tissues. Human population is not uniform in its radiation sensitivity. Radiosensitive sub-groups exist, which would suffer an increased incidence of both deterministic and stochastic effects. Clinical studies have suggested that a large part of the spectrum of normal tissue reaction may be due to differences in individual radiosensitivity. The identification of such sub-groups should be relevant for radiation therapy and radiation protection purposes. It is suggested that DNA repair mechanisms are involved. Consequently, the characterization of DNA repair in lymphocytes through cytokinesis blocked micronucleus (MN) and alkaline single-cell microgel electrophoresis (comet) assays could be a suitable approache to evaluate individual radiosensitivity in vitro. The aims of this study were: 1) To assess the in vitro radiosensitivity of peripheral blood lymphocytes from two groups of cancer patients (prospectively and retrospectively studied), using MN and comet assays, in comparison with the clinical radiation reaction and 2) To test the predictive potential of both techniques for the identification of radiosensitivity sub-groups. 38 cancer patients receiving radiation therapy were enrolled in this study. 19 patients were evaluated prior, mid-way and on completion of treatment (prospective group) and 19 patients were evaluated about 6-18 month after radiotherapy (retrospective group). Cytogenetic data from the prospective group were analysed using a mathematical model to evaluate the attenuation of the cytogenetic effect as a function of the time between a single exposure and blood sampling, estimating a cytogenetic recovery factor k. In the retrospective group, blood samples were irradiated in vitro with 0 (control) or 2 Gy and evaluated using MN test. Cytogenetic data were analysed

  1. Lessons learned during Type A Packaging testing

    International Nuclear Information System (INIS)

    O'Brien, J.H.; Kelly, D.L.

    1995-11-01

    For the past 6 years, the US Department of Energy (DOE) Office of Facility Safety Analysis (EH-32) has contracted Westinghouse Hanford Company (WHC) to conduct compliance testing on DOE Type A packagings. The packagings are tested for compliance with the U.S. Department of Transportation (DOT) Specification 7A, general packaging, Type A requirements. The DOE has shared the Type A packaging information throughout the nuclear materials transportation community. During testing, there have been recurring areas of packaging design that resulted in testing delays and/or initial failure. The lessons learned during the testing are considered a valuable resource. DOE requested that WHC share this resource. By sharing what is and can be encountered during packaging testing, individuals will hopefully avoid past mistakes

  2. WORMHOLE: Novel Least Diverged Ortholog Prediction through Machine Learning.

    Directory of Open Access Journals (Sweden)

    George L Sutphin

    2016-11-01

    Full Text Available The rapid advancement of technology in genomics and targeted genetic manipulation has made comparative biology an increasingly prominent strategy to model human disease processes. Predicting orthology relationships between species is a vital component of comparative biology. Dozens of strategies for predicting orthologs have been developed using combinations of gene and protein sequence, phylogenetic history, and functional interaction with progressively increasing accuracy. A relatively new class of orthology prediction strategies combines aspects of multiple methods into meta-tools, resulting in improved prediction performance. Here we present WORMHOLE, a novel ortholog prediction meta-tool that applies machine learning to integrate 17 distinct ortholog prediction algorithms to identify novel least diverged orthologs (LDOs between 6 eukaryotic species-humans, mice, zebrafish, fruit flies, nematodes, and budding yeast. Machine learning allows WORMHOLE to intelligently incorporate predictions from a wide-spectrum of strategies in order to form aggregate predictions of LDOs with high confidence. In this study we demonstrate the performance of WORMHOLE across each combination of query and target species. We show that WORMHOLE is particularly adept at improving LDO prediction performance between distantly related species, expanding the pool of LDOs while maintaining low evolutionary distance and a high level of functional relatedness between genes in LDO pairs. We present extensive validation, including cross-validated prediction of PANTHER LDOs and evaluation of evolutionary divergence and functional similarity, and discuss future applications of machine learning in ortholog prediction. A WORMHOLE web tool has been developed and is available at http://wormhole.jax.org/.

  3. WORMHOLE: Novel Least Diverged Ortholog Prediction through Machine Learning

    Science.gov (United States)

    Sutphin, George L.; Mahoney, J. Matthew; Sheppard, Keith; Walton, David O.; Korstanje, Ron

    2016-01-01

    The rapid advancement of technology in genomics and targeted genetic manipulation has made comparative biology an increasingly prominent strategy to model human disease processes. Predicting orthology relationships between species is a vital component of comparative biology. Dozens of strategies for predicting orthologs have been developed using combinations of gene and protein sequence, phylogenetic history, and functional interaction with progressively increasing accuracy. A relatively new class of orthology prediction strategies combines aspects of multiple methods into meta-tools, resulting in improved prediction performance. Here we present WORMHOLE, a novel ortholog prediction meta-tool that applies machine learning to integrate 17 distinct ortholog prediction algorithms to identify novel least diverged orthologs (LDOs) between 6 eukaryotic species—humans, mice, zebrafish, fruit flies, nematodes, and budding yeast. Machine learning allows WORMHOLE to intelligently incorporate predictions from a wide-spectrum of strategies in order to form aggregate predictions of LDOs with high confidence. In this study we demonstrate the performance of WORMHOLE across each combination of query and target species. We show that WORMHOLE is particularly adept at improving LDO prediction performance between distantly related species, expanding the pool of LDOs while maintaining low evolutionary distance and a high level of functional relatedness between genes in LDO pairs. We present extensive validation, including cross-validated prediction of PANTHER LDOs and evaluation of evolutionary divergence and functional similarity, and discuss future applications of machine learning in ortholog prediction. A WORMHOLE web tool has been developed and is available at http://wormhole.jax.org/. PMID:27812085

  4. A predictive validity study of the Learning Style Questionnaire (LSQ) using multiple, specific learning criteria

    NARCIS (Netherlands)

    Kappe, F.R.; Boekholt, L.; den Rooyen, C.; van der Flier, H.

    2009-01-01

    Multiple and specific learning criteria were used to examine the predictive validity of the Learning Style Questionnaire (LSQ). Ninety-nine students in a college of higher learning in The Netherlands participated in a naturally occurring field study. The students were categorized into one of four

  5. Dropout Prediction in E-Learning Courses through the Combination of Machine Learning Techniques

    Science.gov (United States)

    Lykourentzou, Ioanna; Giannoukos, Ioannis; Nikolopoulos, Vassilis; Mpardis, George; Loumos, Vassili

    2009-01-01

    In this paper, a dropout prediction method for e-learning courses, based on three popular machine learning techniques and detailed student data, is proposed. The machine learning techniques used are feed-forward neural networks, support vector machines and probabilistic ensemble simplified fuzzy ARTMAP. Since a single technique may fail to…

  6. Deep learning versus traditional machine learning methods for aggregated energy demand prediction

    NARCIS (Netherlands)

    Paterakis, N.G.; Mocanu, E.; Gibescu, M.; Stappers, B.; van Alst, W.

    2018-01-01

    In this paper the more advanced, in comparison with traditional machine learning approaches, deep learning methods are explored with the purpose of accurately predicting the aggregated energy consumption. Despite the fact that a wide range of machine learning methods have been applied to

  7. Universal LD50 predictions using deep learning

    Science.gov (United States)

    NICEATM Predictive Models for Acute Oral Systemic Toxicity LD50 entry Risa R. Sayre (sayre.risa@epa.gov) & Christopher M. Grulke Our approach uses an ensemble of multilayer perceptron regressions to predict rat acute oral LD50 values from chemical features. Features were genera...

  8. Machine learning in Python essential techniques for predictive analysis

    CERN Document Server

    Bowles, Michael

    2015-01-01

    Learn a simpler and more effective way to analyze data and predict outcomes with Python Machine Learning in Python shows you how to successfully analyze data using only two core machine learning algorithms, and how to apply them using Python. By focusing on two algorithm families that effectively predict outcomes, this book is able to provide full descriptions of the mechanisms at work, and the examples that illustrate the machinery with specific, hackable code. The algorithms are explained in simple terms with no complex math and applied using Python, with guidance on algorithm selection, d

  9. Decoding the future from past experience: learning shapes predictions in early visual cortex.

    Science.gov (United States)

    Luft, Caroline D B; Meeson, Alan; Welchman, Andrew E; Kourtzi, Zoe

    2015-05-01

    Learning the structure of the environment is critical for interpreting the current scene and predicting upcoming events. However, the brain mechanisms that support our ability to translate knowledge about scene statistics to sensory predictions remain largely unknown. Here we provide evidence that learning of temporal regularities shapes representations in early visual cortex that relate to our ability to predict sensory events. We tested the participants' ability to predict the orientation of a test stimulus after exposure to sequences of leftward- or rightward-oriented gratings. Using fMRI decoding, we identified brain patterns related to the observers' visual predictions rather than stimulus-driven activity. Decoding of predicted orientations following structured sequences was enhanced after training, while decoding of cued orientations following exposure to random sequences did not change. These predictive representations appear to be driven by the same large-scale neural populations that encode actual stimulus orientation and to be specific to the learned sequence structure. Thus our findings provide evidence that learning temporal structures supports our ability to predict future events by reactivating selective sensory representations as early as in primary visual cortex. Copyright © 2015 the American Physiological Society.

  10. Learning to predict is spared in mild cognitive impairment due to Alzheimer's disease.

    Science.gov (United States)

    Baker, Rosalind; Bentham, Peter; Kourtzi, Zoe

    2015-10-01

    Learning the statistics of the environment is critical for predicting upcoming events. However, little is known about how we translate previous knowledge about scene regularities to sensory predictions. Here, we ask whether patients with mild cognitive impairment due to Alzheimer's disease (MCI-AD) that are known to have spared implicit but impaired explicit recognition memory are able to learn temporal regularities and predict upcoming events. We tested the ability of MCI-AD patients and age-matched controls to predict the orientation of a test stimulus following exposure to sequences of leftwards or rightwards oriented gratings. Our results demonstrate that exposure to temporal sequences without feedback facilitates the ability to predict an upcoming stimulus in both MCI-AD patients and controls. Further, we show that executive cognitive control may account for individual variability in predictive learning. That is, we observed significant positive correlations of performance in attentional and working memory tasks with post-training performance in the prediction task. Taken together, these results suggest a mediating role of circuits involved in cognitive control (i.e. frontal circuits) that may support the ability for predictive learning in MCI-AD.

  11. Boosting compound-protein interaction prediction by deep learning.

    Science.gov (United States)

    Tian, Kai; Shao, Mingyu; Wang, Yang; Guan, Jihong; Zhou, Shuigeng

    2016-11-01

    The identification of interactions between compounds and proteins plays an important role in network pharmacology and drug discovery. However, experimentally identifying compound-protein interactions (CPIs) is generally expensive and time-consuming, computational approaches are thus introduced. Among these, machine-learning based methods have achieved a considerable success. However, due to the nonlinear and imbalanced nature of biological data, many machine learning approaches have their own limitations. Recently, deep learning techniques show advantages over many state-of-the-art machine learning methods in some applications. In this study, we aim at improving the performance of CPI prediction based on deep learning, and propose a method called DL-CPI (the abbreviation of Deep Learning for Compound-Protein Interactions prediction), which employs deep neural network (DNN) to effectively learn the representations of compound-protein pairs. Extensive experiments show that DL-CPI can learn useful features of compound-protein pairs by a layerwise abstraction, and thus achieves better prediction performance than existing methods on both balanced and imbalanced datasets. Copyright © 2016 Elsevier Inc. All rights reserved.

  12. Robust portfolio choice with ambiguity and learning about return predictability

    DEFF Research Database (Denmark)

    Larsen, Linda Sandris; Branger, Nicole; Munk, Claus

    2013-01-01

    We analyze the optimal stock-bond portfolio under both learning and ambiguity aversion. Stock returns are predictable by an observable and an unobservable predictor, and the investor has to learn about the latter. Furthermore, the investor is ambiguity-averse and has a preference for investment...... strategies that are robust to model misspecifications. We derive a closed-form solution for the optimal robust investment strategy. We find that both learning and ambiguity aversion impact the level and structure of the optimal stock investment. Suboptimal strategies resulting either from not learning...... or from not considering ambiguity can lead to economically significant losses....

  13. Predicting Contextual Informativeness for Vocabulary Learning

    Science.gov (United States)

    Kapelner, Adam; Soterwood, Jeanine; Nessaiver, Shalev; Adlof, Suzanne

    2018-01-01

    Vocabulary knowledge is essential to educational progress. High quality vocabulary instruction requires supportive contextual examples to teach word meaning and proper usage. Identifying such contexts by hand for a large number of words can be difficult. In this work, we take a statistical learning approach to engineer a system that predicts…

  14. Improving sequence segmentation learning by predicting trigrams

    NARCIS (Netherlands)

    van den Bosch, A.; Daelemans, W.; Dagan, I.; Gildea, D.

    2005-01-01

    Symbolic machine-learning classifiers are known to suffer from near-sightedness when performing sequence segmentation (chunking) tasks in natural language processing: without special architectural additions they are oblivious of the decisions they made earlier when making new ones. We introduce a

  15. Federated learning of predictive models from federated Electronic Health Records.

    Science.gov (United States)

    Brisimi, Theodora S; Chen, Ruidi; Mela, Theofanie; Olshevsky, Alex; Paschalidis, Ioannis Ch; Shi, Wei

    2018-04-01

    In an era of "big data," computationally efficient and privacy-aware solutions for large-scale machine learning problems become crucial, especially in the healthcare domain, where large amounts of data are stored in different locations and owned by different entities. Past research has been focused on centralized algorithms, which assume the existence of a central data repository (database) which stores and can process the data from all participants. Such an architecture, however, can be impractical when data are not centrally located, it does not scale well to very large datasets, and introduces single-point of failure risks which could compromise the integrity and privacy of the data. Given scores of data widely spread across hospitals/individuals, a decentralized computationally scalable methodology is very much in need. We aim at solving a binary supervised classification problem to predict hospitalizations for cardiac events using a distributed algorithm. We seek to develop a general decentralized optimization framework enabling multiple data holders to collaborate and converge to a common predictive model, without explicitly exchanging raw data. We focus on the soft-margin l 1 -regularized sparse Support Vector Machine (sSVM) classifier. We develop an iterative cluster Primal Dual Splitting (cPDS) algorithm for solving the large-scale sSVM problem in a decentralized fashion. Such a distributed learning scheme is relevant for multi-institutional collaborations or peer-to-peer applications, allowing the data holders to collaborate, while keeping every participant's data private. We test cPDS on the problem of predicting hospitalizations due to heart diseases within a calendar year based on information in the patients Electronic Health Records prior to that year. cPDS converges faster than centralized methods at the cost of some communication between agents. It also converges faster and with less communication overhead compared to an alternative distributed

  16. Monitoring and regulation of learning in medical education: the need for predictive cues.

    Science.gov (United States)

    de Bruin, Anique B H; Dunlosky, John; Cavalcanti, Rodrigo B

    2017-06-01

    Being able to accurately monitor learning activities is a key element in self-regulated learning in all settings, including medical schools. Yet students' ability to monitor their progress is often limited, leading to inefficient use of study time. Interventions that improve the accuracy of students' monitoring can optimise self-regulated learning, leading to higher achievement. This paper reviews findings from cognitive psychology and explores potential applications in medical education, as well as areas for future research. Effective monitoring depends on students' ability to generate information ('cues') that accurately reflects their knowledge and skills. The ability of these 'cues' to predict achievement is referred to as 'cue diagnosticity'. Interventions that improve the ability of students to elicit predictive cues typically fall into two categories: (i) self-generation of cues and (ii) generation of cues that is delayed after self-study. Providing feedback and support is useful when cues are predictive but may be too complex to be readily used. Limited evidence exists about interventions to improve the accuracy of self-monitoring among medical students or trainees. Developing interventions that foster use of predictive cues can enhance the accuracy of self-monitoring, thereby improving self-study and clinical reasoning. First, insight should be gained into the characteristics of predictive cues used by medical students and trainees. Next, predictive cue prompts should be designed and tested to improve monitoring and regulation of learning. Finally, the use of predictive cues should be explored in relation to teaching and learning clinical reasoning. Improving self-regulated learning is important to help medical students and trainees efficiently acquire knowledge and skills necessary for clinical practice. Interventions that help students generate and use predictive cues hold the promise of improved self-regulated learning and achievement. This framework is

  17. Are traditional cognitive tests useful in predicting clinical success?

    Science.gov (United States)

    Gray, Sarah A; Deem, Lisa P; Straja, Sorin R

    2002-11-01

    The purpose of this research was to determine the predictive value of the Dental Admission Test (DAT) for clinical success using Ackerman's theory of ability determinants of skilled performance. The Ackerman theory is a valid, reliable schema in the applied psychology literature used to predict complex skill acquisition. Inconsistent stimulus-response skill acquisition depends primarily on determinants of cognitive ability. Consistent information-processing tasks have been described as "automatic," in which stimuli and responses are mapped in a manner that allows for complete certainty once the relationships have been learned. It is theorized that the skills necessary for success in the clinical component of dental schools involve a significant amount of automatic processing demands and, as such, student performance in the clinics should begin to converge as task practice is realized and tasks become more consistent. Subtest scores of the DAT of four classes were correlated with final grades in nine clinical courses. Results showed that the DAT subtest scores played virtually no role with regard to the final clinical grades. Based on this information, the DAT scores were determined to be of no predictive value in clinical achievement.

  18. Learning to forget: continual prediction with LSTM.

    Science.gov (United States)

    Gers, F A; Schmidhuber, J; Cummins, F

    2000-10-01

    Long short-term memory (LSTM; Hochreiter & Schmidhuber, 1997) can solve numerous tasks not solvable by previous learning algorithms for recurrent neural networks (RNNs). We identify a weakness of LSTM networks processing continual input streams that are not a priori segmented into subsequences with explicitly marked ends at which the network's internal state could be reset. Without resets, the state may grow indefinitely and eventually cause the network to break down. Our remedy is a novel, adaptive "forget gate" that enables an LSTM cell to learn to reset itself at appropriate times, thus releasing internal resources. We review illustrative benchmark problems on which standard LSTM outperforms other RNN algorithms. All algorithms (including LSTM) fail to solve continual versions of these problems. LSTM with forget gates, however, easily solves them, and in an elegant way.

  19. Proactive Interference in Human Predictive Learning

    OpenAIRE

    Castro, Leyre; Ortega, Nuria; Matute, Helena

    2002-01-01

    The impairment in responding to a secondly trained association because of the prior training of another (i.e., proactive interference) is a well-established effect in human and animal research, and it has been demonstrated in many paradigms. However, learning theories have been concerned with proactive interference only when the competing stimuli have been presented in compound at some moment of the training phase. In this experiment we investigated the possibility of proactive interference b...

  20. Can Learning Motivation Predict Learning Achievement? A Case Study of a Mobile Game-Based English Learning Approach

    Science.gov (United States)

    Tsai, Chia-Hui; Cheng, Ching-Hsue; Yeh, Duen-Yian; Lin, Shih-Yun

    2017-01-01

    This study applied a quasi-experimental design to investigate the influence and predictive power of learner motivation for achievement, employing a mobile game-based English learning approach. A system called the Happy English Learning System, integrating learning material into a game-based context, was constructed and installed on mobile devices…

  1. Visual Descriptor Learning for Predicting Grasping Affordances

    DEFF Research Database (Denmark)

    Thomsen, Mikkel Tang

    2016-01-01

    by the task of grasping unknown objects given visual sensor information. The contributions from this thesis stem from three works that all relate to the task of grasping unknown objects but with particular focus on the visual representation part of the problem. First an investigation of a visual feature space...... consisting of surface features was performed. Dimensions in the visual space were varied and the effects were evaluated with the task of grasping unknown object. The evaluation was performed using a novel probabilistic grasp prediction approach based on neighbourhood analysis. The resulting success......-rates for predicting grasps were between 75% and 90% depending on the object class. The investigations also provided insights into the importance of selecting a proper visual feature space when utilising it for predicting affordances. As a consequence of the gained insights, a semi-local surface feature, the Sliced...

  2. Playing off the curve - testing quantitative predictions of skill acquisition theories in development of chess performance.

    Science.gov (United States)

    Gaschler, Robert; Progscha, Johanna; Smallbone, Kieran; Ram, Nilam; Bilalić, Merim

    2014-01-01

    Learning curves have been proposed as an adequate description of learning processes, no matter whether the processes manifest within minutes or across years. Different mechanisms underlying skill acquisition can lead to differences in the shape of learning curves. In the current study, we analyze the tournament performance data of 1383 chess players who begin competing at young age and play tournaments for at least 10 years. We analyze the performance development with the goal to test the adequacy of learning curves, and the skill acquisition theories they are based on, for describing and predicting expertise acquisition. On the one hand, we show that the skill acquisition theories implying a negative exponential learning curve do a better job in both describing early performance gains and predicting later trajectories of chess performance than those theories implying a power function learning curve. On the other hand, the learning curves of a large proportion of players show systematic qualitative deviations from the predictions of either type of skill acquisition theory. While skill acquisition theories predict larger performance gains in early years and smaller gains in later years, a substantial number of players begin to show substantial improvements with a delay of several years (and no improvement in the first years), deviations not fully accounted for by quantity of practice. The current work adds to the debate on how learning processes on a small time scale combine to large-scale changes.

  3. Learning from sensory and reward prediction errors during motor adaptation.

    Science.gov (United States)

    Izawa, Jun; Shadmehr, Reza

    2011-03-01

    Voluntary motor commands produce two kinds of consequences. Initially, a sensory consequence is observed in terms of activity in our primary sensory organs (e.g., vision, proprioception). Subsequently, the brain evaluates the sensory feedback and produces a subjective measure of utility or usefulness of the motor commands (e.g., reward). As a result, comparisons between predicted and observed consequences of motor commands produce two forms of prediction error. How do these errors contribute to changes in motor commands? Here, we considered a reach adaptation protocol and found that when high quality sensory feedback was available, adaptation of motor commands was driven almost exclusively by sensory prediction errors. This form of learning had a distinct signature: as motor commands adapted, the subjects altered their predictions regarding sensory consequences of motor commands, and generalized this learning broadly to neighboring motor commands. In contrast, as the quality of the sensory feedback degraded, adaptation of motor commands became more dependent on reward prediction errors. Reward prediction errors produced comparable changes in the motor commands, but produced no change in the predicted sensory consequences of motor commands, and generalized only locally. Because we found that there was a within subject correlation between generalization patterns and sensory remapping, it is plausible that during adaptation an individual's relative reliance on sensory vs. reward prediction errors could be inferred. We suggest that while motor commands change because of sensory and reward prediction errors, only sensory prediction errors produce a change in the neural system that predicts sensory consequences of motor commands.

  4. Spontaneous brain activity predicts learning ability of foreign sounds.

    Science.gov (United States)

    Ventura-Campos, Noelia; Sanjuán, Ana; González, Julio; Palomar-García, María-Ángeles; Rodríguez-Pujadas, Aina; Sebastián-Gallés, Núria; Deco, Gustavo; Ávila, César

    2013-05-29

    Can learning capacity of the human brain be predicted from initial spontaneous functional connectivity (FC) between brain areas involved in a task? We combined task-related functional magnetic resonance imaging (fMRI) and resting-state fMRI (rs-fMRI) before and after training with a Hindi dental-retroflex nonnative contrast. Previous fMRI results were replicated, demonstrating that this learning recruited the left insula/frontal operculum and the left superior parietal lobe, among other areas of the brain. Crucially, resting-state FC (rs-FC) between these two areas at pretraining predicted individual differences in learning outcomes after distributed (Experiment 1) and intensive training (Experiment 2). Furthermore, this rs-FC was reduced at posttraining, a change that may also account for learning. Finally, resting-state network analyses showed that the mechanism underlying this reduction of rs-FC was mainly a transfer in intrinsic activity of the left frontal operculum/anterior insula from the left frontoparietal network to the salience network. Thus, rs-FC may contribute to predict learning ability and to understand how learning modifies the functioning of the brain. The discovery of this correspondence between initial spontaneous brain activity in task-related areas and posttraining performance opens new avenues to find predictors of learning capacities in the brain using task-related fMRI and rs-fMRI combined.

  5. Regional hippocampal volumes and development predict learning and memory.

    Science.gov (United States)

    Tamnes, Christian K; Walhovd, Kristine B; Engvig, Andreas; Grydeland, Håkon; Krogsrud, Stine K; Østby, Ylva; Holland, Dominic; Dale, Anders M; Fjell, Anders M

    2014-01-01

    The hippocampus is an anatomically and functionally heterogeneous structure, but longitudinal studies of its regional development are scarce and it is not known whether protracted maturation of the hippocampus in adolescence is related to memory development. First, we investigated hippocampal subfield development using 170 longitudinally acquired brain magnetic resonance imaging scans from 85 participants aged 8-21 years. Hippocampal subfield volumes were estimated by the use of automated segmentation of 7 subfields, including the cornu ammonis (CA) sectors and the dentate gyrus (DG), while longitudinal subfield volumetric change was quantified using a nonlinear registration procedure. Second, associations between subfield volumes and change and verbal learning/memory across multiple retention intervals (5 min, 30 min and 1 week) were tested. It was hypothesized that short and intermediate memory would be more closely related to CA2-3/CA4-DG and extended, remote memory to CA1. Change rates were significantly different across hippocampal subfields, but nearly all subfields showed significant volume decreases over time throughout adolescence. Several subfield volumes were larger in the right hemisphere and in males, while for change rates there were no hemisphere or sex differences. Partly in support of the hypotheses, greater volume of CA1 and CA2-3 was related to recall and retention after an extended delay, while longitudinal reduction of CA2-3 and CA4-DG was related to learning. This suggests continued regional development of the hippocampus across adolescence and that volume and volume change in specific subfields differentially predict verbal learning and memory over different retention intervals, but future high-resolution studies are called for. © 2014 S. Karger AG, Basel.

  6. Deep-Learning-Based Approach for Prediction of Algal Blooms

    Directory of Open Access Journals (Sweden)

    Feng Zhang

    2016-10-01

    Full Text Available Algal blooms have recently become a critical global environmental concern which might put economic development and sustainability at risk. However, the accurate prediction of algal blooms remains a challenging scientific problem. In this study, a novel prediction approach for algal blooms based on deep learning is presented—a powerful tool to represent and predict highly dynamic and complex phenomena. The proposed approach constructs a five-layered model to extract detailed relationships between the density of phytoplankton cells and various environmental parameters. The algal blooms can be predicted by the phytoplankton density obtained from the output layer. A case study is conducted in coastal waters of East China using both our model and a traditional back-propagation neural network for comparison. The results show that the deep-learning-based model yields better generalization and greater accuracy in predicting algal blooms than a traditional shallow neural network does.

  7. Spontaneous eye movements and trait empathy predict vicarious learning of fear.

    Science.gov (United States)

    Kleberg, Johan L; Selbing, Ida; Lundqvist, Daniel; Hofvander, Björn; Olsson, Andreas

    2015-12-01

    Learning to predict dangerous outcomes is important to survival. In humans, this kind of learning is often transmitted through the observation of others' emotional responses. We analyzed eye movements during an observational/vicarious fear learning procedure, in which healthy participants (N=33) watched another individual ('learning model') receiving aversive treatment (shocks) paired with a predictive conditioned stimulus (CS+), but not a control stimulus (CS-). Participants' gaze pattern towards the model differentiated as a function of whether the CS was predictive or not of a shock to the model. Consistent with our hypothesis that the face of a conspecific in distress can act as an unconditioned stimulus (US), we found that the total fixation time at a learning model's face increased when the CS+ was shown. Furthermore, we found that the total fixation time at the CS+ during learning predicted participants' conditioned responses (CRs) at a later test in the absence of the model. We also demonstrated that trait empathy was associated with stronger CRs, and that autistic traits were positively related to autonomic reactions to watching the model receiving the aversive treatment. Our results have implications for both healthy and dysfunctional socio-emotional learning. Copyright © 2015 Elsevier B.V. All rights reserved.

  8. Test-enhanced learning: the potential for testing to promote greater learning in undergraduate science courses.

    Science.gov (United States)

    Brame, Cynthia J; Biel, Rachel

    2015-01-01

    Testing within the science classroom is commonly used for both formative and summative assessment purposes to let the student and the instructor gauge progress toward learning goals. Research within cognitive science suggests, however, that testing can also be a learning event. We present summaries of studies that suggest that repeated retrieval can enhance long-term learning in a laboratory setting; various testing formats can promote learning; feedback enhances the benefits of testing; testing can potentiate further study; and benefits of testing are not limited to rote memory. Most of these studies were performed in a laboratory environment, so we also present summaries of experiments suggesting that the benefits of testing can extend to the classroom. Finally, we suggest opportunities that these observations raise for the classroom and for further research. © 2015 C. J. Brame and R. Biel. CBE—Life Sciences Education © 2015 The American Society for Cell Biology. This article is distributed by The American Society for Cell Biology under license from the author(s). It is available to the public under an Attribution–Noncommercial–Share Alike 3.0 Unported Creative Commons License (http://creativecommons.org/licenses/by-nc-sa/3.0).

  9. Learning biases predict a word order universal.

    Science.gov (United States)

    Culbertson, Jennifer; Smolensky, Paul; Legendre, Géraldine

    2012-03-01

    How recurrent typological patterns, or universals, emerge from the extensive diversity found across the world's languages constitutes a central question for linguistics and cognitive science. Recent challenges to a fundamental assumption of generative linguistics-that universal properties of the human language acquisition faculty constrain the types of grammatical systems which can occur-suggest the need for new types of empirical evidence connecting typology to biases of learners. Using an artificial language learning paradigm in which adult subjects are exposed to a mix of grammatical systems (similar to a period of linguistic change), we show that learners' biases mirror a word-order universal, first proposed by Joseph Greenberg, which constrains typological patterns of adjective, numeral, and noun ordering. We briefly summarize the results of a probabilistic model of the hypothesized biases and their effect on learning, and discuss the broader implications of the results for current theories of the origins of cross-linguistic word-order preferences. Copyright © 2011 Elsevier B.V. All rights reserved.

  10. Predicting Market Impact Costs Using Nonparametric Machine Learning Models.

    Directory of Open Access Journals (Sweden)

    Saerom Park

    Full Text Available Market impact cost is the most significant portion of implicit transaction costs that can reduce the overall transaction cost, although it cannot be measured directly. In this paper, we employed the state-of-the-art nonparametric machine learning models: neural networks, Bayesian neural network, Gaussian process, and support vector regression, to predict market impact cost accurately and to provide the predictive model that is versatile in the number of variables. We collected a large amount of real single transaction data of US stock market from Bloomberg Terminal and generated three independent input variables. As a result, most nonparametric machine learning models outperformed a-state-of-the-art benchmark parametric model such as I-star model in four error measures. Although these models encounter certain difficulties in separating the permanent and temporary cost directly, nonparametric machine learning models can be good alternatives in reducing transaction costs by considerably improving in prediction performance.

  11. Predicting Market Impact Costs Using Nonparametric Machine Learning Models.

    Science.gov (United States)

    Park, Saerom; Lee, Jaewook; Son, Youngdoo

    2016-01-01

    Market impact cost is the most significant portion of implicit transaction costs that can reduce the overall transaction cost, although it cannot be measured directly. In this paper, we employed the state-of-the-art nonparametric machine learning models: neural networks, Bayesian neural network, Gaussian process, and support vector regression, to predict market impact cost accurately and to provide the predictive model that is versatile in the number of variables. We collected a large amount of real single transaction data of US stock market from Bloomberg Terminal and generated three independent input variables. As a result, most nonparametric machine learning models outperformed a-state-of-the-art benchmark parametric model such as I-star model in four error measures. Although these models encounter certain difficulties in separating the permanent and temporary cost directly, nonparametric machine learning models can be good alternatives in reducing transaction costs by considerably improving in prediction performance.

  12. Predicting the dissolution kinetics of silicate glasses using machine learning

    Science.gov (United States)

    Anoop Krishnan, N. M.; Mangalathu, Sujith; Smedskjaer, Morten M.; Tandia, Adama; Burton, Henry; Bauchy, Mathieu

    2018-05-01

    Predicting the dissolution rates of silicate glasses in aqueous conditions is a complex task as the underlying mechanism(s) remain poorly understood and the dissolution kinetics can depend on a large number of intrinsic and extrinsic factors. Here, we assess the potential of data-driven models based on machine learning to predict the dissolution rates of various aluminosilicate glasses exposed to a wide range of solution pH values, from acidic to caustic conditions. Four classes of machine learning methods are investigated, namely, linear regression, support vector machine regression, random forest, and artificial neural network. We observe that, although linear methods all fail to describe the dissolution kinetics, the artificial neural network approach offers excellent predictions, thanks to its inherent ability to handle non-linear data. Overall, we suggest that a more extensive use of machine learning approaches could significantly accelerate the design of novel glasses with tailored properties.

  13. Machine learning algorithms for datasets popularity prediction

    CERN Document Server

    Kancys, Kipras

    2016-01-01

    This report represents continued study where ML algorithms were used to predict databases popularity. Three topics were covered. First of all, there was a discrepancy between old and new meta-data collection procedures, so a reason for that had to be found. Secondly, different parameters were analysed and dropped to make algorithms perform better. And third, it was decided to move modelling part on Spark.

  14. A Test of Two Alternative Cognitive Processing Models: Learning Styles and Dual Coding

    Science.gov (United States)

    Cuevas, Joshua; Dawson, Bryan L.

    2018-01-01

    This study tested two cognitive models, learning styles and dual coding, which make contradictory predictions about how learners process and retain visual and auditory information. Learning styles-based instructional practices are common in educational environments despite a questionable research base, while the use of dual coding is less…

  15. Pressure Prediction of Coal Slurry Transportation Pipeline Based on Particle Swarm Optimization Kernel Function Extreme Learning Machine

    Directory of Open Access Journals (Sweden)

    Xue-cun Yang

    2015-01-01

    Full Text Available For coal slurry pipeline blockage prediction problem, through the analysis of actual scene, it is determined that the pressure prediction from each measuring point is the premise of pipeline blockage prediction. Kernel function of support vector machine is introduced into extreme learning machine, the parameters are optimized by particle swarm algorithm, and blockage prediction method based on particle swarm optimization kernel function extreme learning machine (PSOKELM is put forward. The actual test data from HuangLing coal gangue power plant are used for simulation experiments and compared with support vector machine prediction model optimized by particle swarm algorithm (PSOSVM and kernel function extreme learning machine prediction model (KELM. The results prove that mean square error (MSE for the prediction model based on PSOKELM is 0.0038 and the correlation coefficient is 0.9955, which is superior to prediction model based on PSOSVM in speed and accuracy and superior to KELM prediction model in accuracy.

  16. Learning predictive statistics from temporal sequences: Dynamics and strategies.

    Science.gov (United States)

    Wang, Rui; Shen, Yuan; Tino, Peter; Welchman, Andrew E; Kourtzi, Zoe

    2017-10-01

    Human behavior is guided by our expectations about the future. Often, we make predictions by monitoring how event sequences unfold, even though such sequences may appear incomprehensible. Event structures in the natural environment typically vary in complexity, from simple repetition to complex probabilistic combinations. How do we learn these structures? Here we investigate the dynamics of structure learning by tracking human responses to temporal sequences that change in structure unbeknownst to the participants. Participants were asked to predict the upcoming item following a probabilistic sequence of symbols. Using a Markov process, we created a family of sequences, from simple frequency statistics (e.g., some symbols are more probable than others) to context-based statistics (e.g., symbol probability is contingent on preceding symbols). We demonstrate the dynamics with which individuals adapt to changes in the environment's statistics-that is, they extract the behaviorally relevant structures to make predictions about upcoming events. Further, we show that this structure learning relates to individual decision strategy; faster learning of complex structures relates to selection of the most probable outcome in a given context (maximizing) rather than matching of the exact sequence statistics. Our findings provide evidence for alternate routes to learning of behaviorally relevant statistics that facilitate our ability to predict future events in variable environments.

  17. Learning new gait patterns: Exploratory muscle activity during motor learning is not predicted by motor modules

    Science.gov (United States)

    Ranganathan, Rajiv; Krishnan, Chandramouli; Dhaher, Yasin Y.; Rymer, William Z.

    2018-01-01

    The motor module hypothesis in motor control proposes that the nervous system can simplify the problem of controlling a large number of muscles in human movement by grouping muscles into a smaller number of modules. Here, we tested one prediction of the modular organization hypothesis by examining whether there is preferential exploration along these motor modules during the learning of a new gait pattern. Healthy college-aged participants learned a new gait pattern which required increased hip and knee flexion during the swing phase while walking in a lower-extremity robot (Lokomat). The new gait pattern was displayed as a foot trajectory in the sagittal plane and participants attempted to match their foot trajectory to this template. We recorded EMG from 8 lower-extremity muscles and we extracted motor modules during both baseline walking and target-tracking using non-negative matrix factorization (NMF). Results showed increased trajectory variability in the first block of learning, indicating that participants were engaged in exploratory behavior. Critically, when we examined the muscle activity during this exploratory phase, we found that the composition of motor modules changed significantly within the first few strides of attempting the new gait pattern. The lack of persistence of the motor modules under even short time scales suggests that motor modules extracted during locomotion may be more indicative of correlated muscle activity induced by the task constraints of walking, rather than reflecting a modular control strategy. PMID:26916510

  18. Just do it: action-dependent learning allows sensory prediction.

    Directory of Open Access Journals (Sweden)

    Itai Novick

    Full Text Available Sensory-motor learning is commonly considered as a mapping process, whereby sensory information is transformed into the motor commands that drive actions. However, this directional mapping, from inputs to outputs, is part of a loop; sensory stimuli cause actions and vice versa. Here, we explore whether actions affect the understanding of the sensory input that they cause. Using a visuo-motor task in humans, we demonstrate two types of learning-related behavioral effects. Stimulus-dependent effects reflect stimulus-response learning, while action-dependent effects reflect a distinct learning component, allowing the brain to predict the forthcoming sensory outcome of actions. Together, the stimulus-dependent and the action-dependent learning components allow the brain to construct a complete internal representation of the sensory-motor loop.

  19. Social learning through prediction error in the brain

    Science.gov (United States)

    Joiner, Jessica; Piva, Matthew; Turrin, Courtney; Chang, Steve W. C.

    2017-06-01

    Learning about the world is critical to survival and success. In social animals, learning about others is a necessary component of navigating the social world, ultimately contributing to increasing evolutionary fitness. How humans and nonhuman animals represent the internal states and experiences of others has long been a subject of intense interest in the developmental psychology tradition, and, more recently, in studies of learning and decision making involving self and other. In this review, we explore how psychology conceptualizes the process of representing others, and how neuroscience has uncovered correlates of reinforcement learning signals to explore the neural mechanisms underlying social learning from the perspective of representing reward-related information about self and other. In particular, we discuss self-referenced and other-referenced types of reward prediction errors across multiple brain structures that effectively allow reinforcement learning algorithms to mediate social learning. Prediction-based computational principles in the brain may be strikingly conserved between self-referenced and other-referenced information.

  20. Predicting Solar Activity Using Machine-Learning Methods

    Science.gov (United States)

    Bobra, M.

    2017-12-01

    Of all the activity observed on the Sun, two of the most energetic events are flares and coronal mass ejections. However, we do not, as of yet, fully understand the physical mechanism that triggers solar eruptions. A machine-learning algorithm, which is favorable in cases where the amount of data is large, is one way to [1] empirically determine the signatures of this mechanism in solar image data and [2] use them to predict solar activity. In this talk, we discuss the application of various machine learning algorithms - specifically, a Support Vector Machine, a sparse linear regression (Lasso), and Convolutional Neural Network - to image data from the photosphere, chromosphere, transition region, and corona taken by instruments aboard the Solar Dynamics Observatory in order to predict solar activity on a variety of time scales. Such an approach may be useful since, at the present time, there are no physical models of flares available for real-time prediction. We discuss our results (Bobra and Couvidat, 2015; Bobra and Ilonidis, 2016; Jonas et al., 2017) as well as other attempts to predict flares using machine-learning (e.g. Ahmed et al., 2013; Nishizuka et al. 2017) and compare these results with the more traditional techniques used by the NOAA Space Weather Prediction Center (Crown, 2012). We also discuss some of the challenges in using machine-learning algorithms for space science applications.

  1. A Model for Predicting Learning Flow and Achievement in Corporate e-Learning

    Science.gov (United States)

    Joo, Young Ju; Lim, Kyu Yon; Kim, Su Mi

    2012-01-01

    The primary objective of this study was to investigate the determinants of learning flow and achievement in corporate online training. Self-efficacy, intrinsic value, and test anxiety were selected as learners' motivational factors, while perceived usefulness and ease of use were also selected as learning environmental factors. Learning flow was…

  2. Neural correlates of testing effects in vocabulary learning.

    Science.gov (United States)

    van den Broek, Gesa S E; Takashima, Atsuko; Segers, Eliane; Fernández, Guillén; Verhoeven, Ludo

    2013-09-01

    Tests that require memory retrieval strongly improve long-term retention in comparison to continued studying. For example, once learners know the translation of a word, restudy practice, during which they see the word and its translation again, is less effective than testing practice, during which they see only the word and retrieve the translation from memory. In the present functional magnetic resonance imaging (fMRI) study, we investigated the neuro-cognitive mechanisms underlying this striking testing effect. Twenty-six young adults without prior knowledge of Swahili learned the translation of 100 Swahili words and then further practiced the words in an fMRI scanner by restudying or by testing. Recall of the translations on a final memory test after one week was significantly better and faster for tested words than for restudied words. Brain regions that were more active during testing than during restudying included the left inferior frontal gyrus, ventral striatum, and midbrain areas. Increased activity in the left inferior parietal and left middle temporal areas during testing but not during restudying predicted better recall on the final memory test. Together, results suggest that testing may be more beneficial than restudying due to processes related to targeted semantic elaboration and selective strengthening of associations between retrieval cues and relevant responses, and may involve increased effortful cognitive control and modulations of memory through striatal motivation and reward circuits. Copyright © 2013 Elsevier Inc. All rights reserved.

  3. Long-term prediction of chaotic time series with multi-step prediction horizons by a neural network with Levenberg-Marquardt learning algorithm

    International Nuclear Information System (INIS)

    Mirzaee, Hossein

    2009-01-01

    The Levenberg-Marquardt learning algorithm is applied for training a multilayer perception with three hidden layer each with ten neurons in order to carefully map the structure of chaotic time series such as Mackey-Glass time series. First the MLP network is trained with 1000 data, and then it is tested with next 500 data. After that the trained and tested network is applied for long-term prediction of next 120 data which come after test data. The prediction is such a way that, the first inputs to network for prediction are the four last data of test data, then the predicted value is shifted to the regression vector which is the input to the network, then after first four-step of prediction, the input regression vector to network is fully predicted values and in continue, each predicted data is shifted to input vector for subsequent prediction.

  4. A Deep Learning Network Approach to ab initio Protein Secondary Structure Prediction.

    Science.gov (United States)

    Spencer, Matt; Eickholt, Jesse; Jianlin Cheng

    2015-01-01

    Ab initio protein secondary structure (SS) predictions are utilized to generate tertiary structure predictions, which are increasingly demanded due to the rapid discovery of proteins. Although recent developments have slightly exceeded previous methods of SS prediction, accuracy has stagnated around 80 percent and many wonder if prediction cannot be advanced beyond this ceiling. Disciplines that have traditionally employed neural networks are experimenting with novel deep learning techniques in attempts to stimulate progress. Since neural networks have historically played an important role in SS prediction, we wanted to determine whether deep learning could contribute to the advancement of this field as well. We developed an SS predictor that makes use of the position-specific scoring matrix generated by PSI-BLAST and deep learning network architectures, which we call DNSS. Graphical processing units and CUDA software optimize the deep network architecture and efficiently train the deep networks. Optimal parameters for the training process were determined, and a workflow comprising three separately trained deep networks was constructed in order to make refined predictions. This deep learning network approach was used to predict SS for a fully independent test dataset of 198 proteins, achieving a Q3 accuracy of 80.7 percent and a Sov accuracy of 74.2 percent.

  5. Time-sensitive Customer Churn Prediction based on PU Learning

    OpenAIRE

    Wang, Li; Chen, Chaochao; Zhou, Jun; Li, Xiaolong

    2018-01-01

    With the fast development of Internet companies throughout the world, customer churn has become a serious concern. To better help the companies retain their customers, it is important to build a customer churn prediction model to identify the customers who are most likely to churn ahead of time. In this paper, we propose a Time-sensitive Customer Churn Prediction (TCCP) framework based on Positive and Unlabeled (PU) learning technique. Specifically, we obtain the recent data by shortening the...

  6. Prediction of Employee Turnover in Organizations using Machine Learning Algorithms

    OpenAIRE

    Rohit Punnoose; Pankaj Ajit

    2016-01-01

    Employee turnover has been identified as a key issue for organizations because of its adverse impact on work place productivity and long term growth strategies. To solve this problem, organizations use machine learning techniques to predict employee turnover. Accurate predictions enable organizations to take action for retention or succession planning of employees. However, the data for this modeling problem comes from HR Information Systems (HRIS); these are typically under-funded compared t...

  7. LESSONS LEARNED IN TESTING OF SAFEGUARDS EQUIPMENT

    International Nuclear Information System (INIS)

    Pepper, S.; Farnitano, M.; Carelli, J.; Hazeltine, J.; Bailey, D.

    2001-01-01

    The International Atomic Energy Agency's (IAEA) Department of Safeguards uses complex instrumentation for the application of safeguards at nuclear facilities around the world. Often, this equipment is developed through cooperation with member state support programs because the Agency's requirements are unique and are not met by commercially available equipment. Before approving an instrument or system for routine inspection use, the IAEA subjects it to a series of tests designed to evaluate its reliability. In 2000, the IAEA began to observe operational failures in digital surveillance systems. In response to the observed failures, the IAEA worked with the equipment designer and manufacturer to determine the cause of failure. An action plan was developed to correct the performance issues and further test the systems to make sure that additional operational issues would not surface later. This paper addresses the steps taken to address operation issues related to digital image surveillance systems and the lessons learned during this process

  8. Predictive ability of machine learning methods for massive crop yield prediction

    Directory of Open Access Journals (Sweden)

    Alberto Gonzalez-Sanchez

    2014-04-01

    Full Text Available An important issue for agricultural planning purposes is the accurate yield estimation for the numerous crops involved in the planning. Machine learning (ML is an essential approach for achieving practical and effective solutions for this problem. Many comparisons of ML methods for yield prediction have been made, seeking for the most accurate technique. Generally, the number of evaluated crops and techniques is too low and does not provide enough information for agricultural planning purposes. This paper compares the predictive accuracy of ML and linear regression techniques for crop yield prediction in ten crop datasets. Multiple linear regression, M5-Prime regression trees, perceptron multilayer neural networks, support vector regression and k-nearest neighbor methods were ranked. Four accuracy metrics were used to validate the models: the root mean square error (RMS, root relative square error (RRSE, normalized mean absolute error (MAE, and correlation factor (R. Real data of an irrigation zone of Mexico were used for building the models. Models were tested with samples of two consecutive years. The results show that M5-Prime and k-nearest neighbor techniques obtain the lowest average RMSE errors (5.14 and 4.91, the lowest RRSE errors (79.46% and 79.78%, the lowest average MAE errors (18.12% and 19.42%, and the highest average correlation factors (0.41 and 0.42. Since M5-Prime achieves the largest number of crop yield models with the lowest errors, it is a very suitable tool for massive crop yield prediction in agricultural planning.

  9. SU-E-J-191: Motion Prediction Using Extreme Learning Machine in Image Guided Radiotherapy

    International Nuclear Information System (INIS)

    Jia, J; Cao, R; Pei, X; Wang, H; Hu, L

    2015-01-01

    Purpose: Real-time motion tracking is a critical issue in image guided radiotherapy due to the time latency caused by image processing and system response. It is of great necessity to fast and accurately predict the future position of the respiratory motion and the tumor location. Methods: The prediction of respiratory position was done based on the positioning and tracking module in ARTS-IGRT system which was developed by FDS Team (www.fds.org.cn). An approach involving with the extreme learning machine (ELM) was adopted to predict the future respiratory position as well as the tumor’s location by training the past trajectories. For the training process, a feed-forward neural network with one single hidden layer was used for the learning. First, the number of hidden nodes was figured out for the single layered feed forward network (SLFN). Then the input weights and hidden layer biases of the SLFN were randomly assigned to calculate the hidden neuron output matrix. Finally, the predicted movement were obtained by applying the output weights and compared with the actual movement. Breathing movement acquired from the external infrared markers was used to test the prediction accuracy. And the implanted marker movement for the prostate cancer was used to test the implementation of the tumor motion prediction. Results: The accuracy of the predicted motion and the actual motion was tested. Five volunteers with different breathing patterns were tested. The average prediction time was 0.281s. And the standard deviation of prediction accuracy was 0.002 for the respiratory motion and 0.001 for the tumor motion. Conclusion: The extreme learning machine method can provide an accurate and fast prediction of the respiratory motion and the tumor location and therefore can meet the requirements of real-time tumor-tracking in image guided radiotherapy

  10. SU-E-J-191: Motion Prediction Using Extreme Learning Machine in Image Guided Radiotherapy

    Energy Technology Data Exchange (ETDEWEB)

    Jia, J; Cao, R; Pei, X; Wang, H; Hu, L [Key Laboratory of Neutronics and Radiation Safety, Institute of Nuclear Energy Safety Technology, Chinese Academy of Sciences, Hefei, Anhui, 230031 (China); Engineering Technology Research Center of Accurate Radiotherapy of Anhui Province, Hefei 230031 (China); Collaborative Innovation Center of Radiation Medicine of Jiangsu Higher Education Institutions, SuZhou (China)

    2015-06-15

    Purpose: Real-time motion tracking is a critical issue in image guided radiotherapy due to the time latency caused by image processing and system response. It is of great necessity to fast and accurately predict the future position of the respiratory motion and the tumor location. Methods: The prediction of respiratory position was done based on the positioning and tracking module in ARTS-IGRT system which was developed by FDS Team (www.fds.org.cn). An approach involving with the extreme learning machine (ELM) was adopted to predict the future respiratory position as well as the tumor’s location by training the past trajectories. For the training process, a feed-forward neural network with one single hidden layer was used for the learning. First, the number of hidden nodes was figured out for the single layered feed forward network (SLFN). Then the input weights and hidden layer biases of the SLFN were randomly assigned to calculate the hidden neuron output matrix. Finally, the predicted movement were obtained by applying the output weights and compared with the actual movement. Breathing movement acquired from the external infrared markers was used to test the prediction accuracy. And the implanted marker movement for the prostate cancer was used to test the implementation of the tumor motion prediction. Results: The accuracy of the predicted motion and the actual motion was tested. Five volunteers with different breathing patterns were tested. The average prediction time was 0.281s. And the standard deviation of prediction accuracy was 0.002 for the respiratory motion and 0.001 for the tumor motion. Conclusion: The extreme learning machine method can provide an accurate and fast prediction of the respiratory motion and the tumor location and therefore can meet the requirements of real-time tumor-tracking in image guided radiotherapy.

  11. Predicting Knowledge Workers' Participation in Voluntary Learning with Employee Characteristics and Online Learning Tools

    Science.gov (United States)

    Hicks, Catherine

    2018-01-01

    Purpose: This paper aims to explore predicting employee learning activity via employee characteristics and usage for two online learning tools. Design/methodology/approach: Statistical analysis focused on observational data collected from user logs. Data are analyzed via regression models. Findings: Findings are presented for over 40,000…

  12. Action-outcome learning and prediction shape the window of simultaneity of audiovisual outcomes.

    Science.gov (United States)

    Desantis, Andrea; Haggard, Patrick

    2016-08-01

    To form a coherent representation of the objects around us, the brain must group the different sensory features composing these objects. Here, we investigated whether actions contribute in this grouping process. In particular, we assessed whether action-outcome learning and prediction contribute to audiovisual temporal binding. Participants were presented with two audiovisual pairs: one pair was triggered by a left action, and the other by a right action. In a later test phase, the audio and visual components of these pairs were presented at different onset times. Participants judged whether they were simultaneous or not. To assess the role of action-outcome prediction on audiovisual simultaneity, each action triggered either the same audiovisual pair as in the learning phase ('predicted' pair), or the pair that had previously been associated with the other action ('unpredicted' pair). We found the time window within which auditory and visual events appeared simultaneous increased for predicted compared to unpredicted pairs. However, no change in audiovisual simultaneity was observed when audiovisual pairs followed visual cues, rather than voluntary actions. This suggests that only action-outcome learning promotes temporal grouping of audio and visual effects. In a second experiment we observed that changes in audiovisual simultaneity do not only depend on our ability to predict what outcomes our actions generate, but also on learning the delay between the action and the multisensory outcome. When participants learned that the delay between action and audiovisual pair was variable, the window of audiovisual simultaneity for predicted pairs increased, relative to a fixed action-outcome pair delay. This suggests that participants learn action-based predictions of audiovisual outcome, and adapt their temporal perception of outcome events based on such predictions. Copyright © 2016 The Authors. Published by Elsevier B.V. All rights reserved.

  13. Learning Political Science with Prediction Markets: An Experimental Study

    Science.gov (United States)

    Ellis, Cali Mortenson; Sami, Rahul

    2012-01-01

    Prediction markets are designed to aggregate the information of many individuals to forecast future events. These markets provide participants with an incentive to seek information and a forum for interaction, making markets a promising tool to motivate student learning. We carried out a quasi-experiment in an introductory political science class…

  14. Protein-Based Urine Test Predicts Kidney Transplant Outcomes

    Science.gov (United States)

    ... News Releases News Release Thursday, August 22, 2013 Protein-based urine test predicts kidney transplant outcomes NIH- ... supporting development of noninvasive tests. Levels of a protein in the urine of kidney transplant recipients can ...

  15. Hebbian Learning is about contingency, not contiguity, and explains the emergence of predictive mirror neurons.

    Science.gov (United States)

    Keysers, Christian; Perrett, David I; Gazzola, Valeria

    2014-04-01

    Hebbian Learning should not be reduced to contiguity, as it detects contingency and causality. Hebbian Learning accounts of mirror neurons make predictions that differ from associative learning: Through Hebbian Learning, mirror neurons become dynamic networks that calculate predictions and prediction errors and relate to ideomotor theories. The social force of imitation is important for mirror neuron emergence and suggests canalization.

  16. Hebbian Learning is about contingency, not contiguity, and explains the emergence of predictive mirror neurons

    OpenAIRE

    Keysers, C.; Perrett, D.I.; Gazzola, V.

    2014-01-01

    Hebbian Learning should not be reduced to contiguity, as it detects contingency and causality. Hebbian Learning accounts of mirror neurons make predictions that differ from associative learning: Through Hebbian Learning, mirror neurons become dynamic networks that calculate predictions and prediction errors and relate to ideomotor theories. The social force of imitation is important for mirror neuron emergence and suggests canalization. Publisher PDF Peer reviewed

  17. Ensemble learned vaccination uptake prediction using web search queries

    DEFF Research Database (Denmark)

    Hansen, Niels Dalum; Lioma, Christina; Mølbak, Kåre

    2016-01-01

    We present a method that uses ensemble learning to combine clinical and web-mined time-series data in order to predict future vaccination uptake. The clinical data is official vaccination registries, and the web data is query frequencies collected from Google Trends. Experiments with official...... vaccine records show that our method predicts vaccination uptake eff?ectively (4.7 Root Mean Squared Error). Whereas performance is best when combining clinical and web data, using solely web data yields comparative performance. To our knowledge, this is the ?first study to predict vaccination uptake...

  18. Prediction of length-of-day using extreme learning machine

    Directory of Open Access Journals (Sweden)

    Yu Lei

    2015-03-01

    Full Text Available Traditional artificial neural networks (ANN such as back-propagation neural networks (BPNN provide good predictions of length-of-day (LOD. However, the determination of network topology is difficult and time consuming. Therefore, we propose a new type of neural network, extreme learning machine (ELM, to improve the efficiency of LOD predictions. Earth orientation parameters (EOP C04 time-series provides daily values from International Earth Rotation and Reference Systems Service (IERS, which serves as our database. First, the known predictable effects that can be described by functional models—such as the effects of solid earth, ocean tides, or seasonal atmospheric variations—are removed a priori from the C04 time-series. Only the residuals after the subtraction of a priori model from the observed LOD data (i.e., the irregular and quasi-periodic variations are employed for training and predictions. The predicted LOD is the sum of a prior extrapolation model and the ELM predictions of the residuals. Different input patterns are discussed and compared to optimize the network solution. The prediction results are analyzed and compared with those obtained by other machine learning-based prediction methods, including BPNN, generalization regression neural networks (GRNN, and adaptive network-based fuzzy inference systems (ANFIS. It is shown that while achieving similar prediction accuracy, the developed method uses much less training time than other methods. Furthermore, to conduct a direct comparison with the existing prediction techniques, the mean-absolute-error (MAE from the proposed method is compared with that from the EOP prediction comparison campaign (EOP PCC. The results indicate that the accuracy of the proposed method is comparable with that of the former techniques. The implementation of the proposed method is simple.

  19. Machine learning-based methods for prediction of linear B-cell epitopes.

    Science.gov (United States)

    Wang, Hsin-Wei; Pai, Tun-Wen

    2014-01-01

    B-cell epitope prediction facilitates immunologists in designing peptide-based vaccine, diagnostic test, disease prevention, treatment, and antibody production. In comparison with T-cell epitope prediction, the performance of variable length B-cell epitope prediction is still yet to be satisfied. Fortunately, due to increasingly available verified epitope databases, bioinformaticians could adopt machine learning-based algorithms on all curated data to design an improved prediction tool for biomedical researchers. Here, we have reviewed related epitope prediction papers, especially those for linear B-cell epitope prediction. It should be noticed that a combination of selected propensity scales and statistics of epitope residues with machine learning-based tools formulated a general way for constructing linear B-cell epitope prediction systems. It is also observed from most of the comparison results that the kernel method of support vector machine (SVM) classifier outperformed other machine learning-based approaches. Hence, in this chapter, except reviewing recently published papers, we have introduced the fundamentals of B-cell epitope and SVM techniques. In addition, an example of linear B-cell prediction system based on physicochemical features and amino acid combinations is illustrated in details.

  20. Deep Learning Accurately Predicts Estrogen Receptor Status in Breast Cancer Metabolomics Data.

    Science.gov (United States)

    Alakwaa, Fadhl M; Chaudhary, Kumardeep; Garmire, Lana X

    2018-01-05

    Metabolomics holds the promise as a new technology to diagnose highly heterogeneous diseases. Conventionally, metabolomics data analysis for diagnosis is done using various statistical and machine learning based classification methods. However, it remains unknown if deep neural network, a class of increasingly popular machine learning methods, is suitable to classify metabolomics data. Here we use a cohort of 271 breast cancer tissues, 204 positive estrogen receptor (ER+), and 67 negative estrogen receptor (ER-) to test the accuracies of feed-forward networks, a deep learning (DL) framework, as well as six widely used machine learning models, namely random forest (RF), support vector machines (SVM), recursive partitioning and regression trees (RPART), linear discriminant analysis (LDA), prediction analysis for microarrays (PAM), and generalized boosted models (GBM). DL framework has the highest area under the curve (AUC) of 0.93 in classifying ER+/ER- patients, compared to the other six machine learning algorithms. Furthermore, the biological interpretation of the first hidden layer reveals eight commonly enriched significant metabolomics pathways (adjusted P-value learning methods. Among them, protein digestion and absorption and ATP-binding cassette (ABC) transporters pathways are also confirmed in integrated analysis between metabolomics and gene expression data in these samples. In summary, deep learning method shows advantages for metabolomics based breast cancer ER status classification, with both the highest prediction accuracy (AUC = 0.93) and better revelation of disease biology. We encourage the adoption of feed-forward networks based deep learning method in the metabolomics research community for classification.

  1. Neural prediction errors reveal a risk-sensitive reinforcement-learning process in the human brain.

    Science.gov (United States)

    Niv, Yael; Edlund, Jeffrey A; Dayan, Peter; O'Doherty, John P

    2012-01-11

    Humans and animals are exquisitely, though idiosyncratically, sensitive to risk or variance in the outcomes of their actions. Economic, psychological, and neural aspects of this are well studied when information about risk is provided explicitly. However, we must normally learn about outcomes from experience, through trial and error. Traditional models of such reinforcement learning focus on learning about the mean reward value of cues and ignore higher order moments such as variance. We used fMRI to test whether the neural correlates of human reinforcement learning are sensitive to experienced risk. Our analysis focused on anatomically delineated regions of a priori interest in the nucleus accumbens, where blood oxygenation level-dependent (BOLD) signals have been suggested as correlating with quantities derived from reinforcement learning. We first provide unbiased evidence that the raw BOLD signal in these regions corresponds closely to a reward prediction error. We then derive from this signal the learned values of cues that predict rewards of equal mean but different variance and show that these values are indeed modulated by experienced risk. Moreover, a close neurometric-psychometric coupling exists between the fluctuations of the experience-based evaluations of risky options that we measured neurally and the fluctuations in behavioral risk aversion. This suggests that risk sensitivity is integral to human learning, illuminating economic models of choice, neuroscientific models of affective learning, and the workings of the underlying neural mechanisms.

  2. Extremely Randomized Machine Learning Methods for Compound Activity Prediction

    Directory of Open Access Journals (Sweden)

    Wojciech M. Czarnecki

    2015-11-01

    Full Text Available Speed, a relatively low requirement for computational resources and high effectiveness of the evaluation of the bioactivity of compounds have caused a rapid growth of interest in the application of machine learning methods to virtual screening tasks. However, due to the growth of the amount of data also in cheminformatics and related fields, the aim of research has shifted not only towards the development of algorithms of high predictive power but also towards the simplification of previously existing methods to obtain results more quickly. In the study, we tested two approaches belonging to the group of so-called ‘extremely randomized methods’—Extreme Entropy Machine and Extremely Randomized Trees—for their ability to properly identify compounds that have activity towards particular protein targets. These methods were compared with their ‘non-extreme’ competitors, i.e., Support Vector Machine and Random Forest. The extreme approaches were not only found out to improve the efficiency of the classification of bioactive compounds, but they were also proved to be less computationally complex, requiring fewer steps to perform an optimization procedure.

  3. Predicting Refractive Surgery Outcome: Machine Learning Approach With Big Data.

    Science.gov (United States)

    Achiron, Asaf; Gur, Zvi; Aviv, Uri; Hilely, Assaf; Mimouni, Michael; Karmona, Lily; Rokach, Lior; Kaiserman, Igor

    2017-09-01

    To develop a decision forest for prediction of laser refractive surgery outcome. Data from consecutive cases of patients who underwent LASIK or photorefractive surgeries during a 12-year period in a single center were assembled into a single dataset. Training of machine-learning classifiers and testing were performed with a statistical classifier algorithm. The decision forest was created by feature vectors extracted from 17,592 cases and 38 clinical parameters for each patient. A 10-fold cross-validation procedure was applied to estimate the predictive value of the decision forest when applied to new patients. Analysis included patients younger than 40 years who were not treated for monovision. Efficacy of 0.7 or greater and 0.8 or greater was achieved in 16,198 (92.0%) and 14,945 (84.9%) eyes, respectively. Efficacy of less than 0.4 and less than 0.5 was achieved in 322 (1.8%) and 506 (2.9%) eyes, respectively. Patients in the low efficacy group (differences compared with the high efficacy group (≥ 0.8), yet were clinically similar (mean differences between groups of 0.7 years, of 0.43 mm in pupil size, of 0.11 D in cylinder, of 0.22 logMAR in preoperative CDVA, of 0.11 mm in optical zone size, of 1.03 D in actual sphere treatment, and of 0.64 D in actual cylinder treatment). The preoperative subjective CDVA had the highest gain (most important to the model). Correlations analysis revealed significantly decreased efficacy with increased age (r = -0.67, P big data from refractive surgeries may be of interest. [J Refract Surg. 2017;33(9):592-597.]. Copyright 2017, SLACK Incorporated.

  4. Predictive genetic tests: problems and pitfalls.

    Science.gov (United States)

    Davis, J G

    1997-12-29

    The role that genetic factors play in medicine has expanded, owing to such recent advances as those made by the Human Genome Project and the work that has spun off from it. The project is focusing particularly on localization and characterization of recognized human genetic disorders, which in turn increases awareness of the potential for improved treatment of these disorders. Technical advances in genetic testing in the absence of effective treatment has presented the health profession with major ethical challenges. The example of the identification of the BRCA1 and BRCA2 genes in families at high risk for breast and ovarian cancer is presented to illustrate the issues of the sensitivity of the method, the degree of susceptibility a positive result implies, the need for and availability of counseling and patient education, and confidentiality of the test results. A compelling need exists for adequate education about medical genetics to raise the "literacy" rate among health professionals.

  5. Students' views of cooperative learning and group testing.

    Science.gov (United States)

    Hicks, Jay

    2007-01-01

    Today's radiologic technology students must learn to collaborate and communicate to function as part of the health care team. Innovative educational techniques such as cooperative learning (working collectively in small groups) and group testing (collaborating on tests) can foster these skills. Assess students' familiarity with and opinions about cooperative learning and group testing before and after participation in a semester-long course incorporating these methods. Twenty-eight students enrolled in a baccalaureate-level radiologic technology program in Louisiana were surveyed at the beginning and end of the semester. Results showed that students were more knowledgeable about and more accepting of cooperative learning and group testing after participating in the course. However, some students continued to prefer independent learning. Students are open to new learning methods such as cooperative learning and group testing. These techniques can help them develop the skills they will need to function collaboratively in the workplace.

  6. How we learn to make decisions: rapid propagation of reinforcement learning prediction errors in humans.

    Science.gov (United States)

    Krigolson, Olav E; Hassall, Cameron D; Handy, Todd C

    2014-03-01

    Our ability to make decisions is predicated upon our knowledge of the outcomes of the actions available to us. Reinforcement learning theory posits that actions followed by a reward or punishment acquire value through the computation of prediction errors-discrepancies between the predicted and the actual reward. A multitude of neuroimaging studies have demonstrated that rewards and punishments evoke neural responses that appear to reflect reinforcement learning prediction errors [e.g., Krigolson, O. E., Pierce, L. J., Holroyd, C. B., & Tanaka, J. W. Learning to become an expert: Reinforcement learning and the acquisition of perceptual expertise. Journal of Cognitive Neuroscience, 21, 1833-1840, 2009; Bayer, H. M., & Glimcher, P. W. Midbrain dopamine neurons encode a quantitative reward prediction error signal. Neuron, 47, 129-141, 2005; O'Doherty, J. P. Reward representations and reward-related learning in the human brain: Insights from neuroimaging. Current Opinion in Neurobiology, 14, 769-776, 2004; Holroyd, C. B., & Coles, M. G. H. The neural basis of human error processing: Reinforcement learning, dopamine, and the error-related negativity. Psychological Review, 109, 679-709, 2002]. Here, we used the brain ERP technique to demonstrate that not only do rewards elicit a neural response akin to a prediction error but also that this signal rapidly diminished and propagated to the time of choice presentation with learning. Specifically, in a simple, learnable gambling task, we show that novel rewards elicited a feedback error-related negativity that rapidly decreased in amplitude with learning. Furthermore, we demonstrate the existence of a reward positivity at choice presentation, a previously unreported ERP component that has a similar timing and topography as the feedback error-related negativity that increased in amplitude with learning. The pattern of results we observed mirrored the output of a computational model that we implemented to compute reward

  7. Historical maintenance relevant information road-map for a self-learning maintenance prediction procedural approach

    Science.gov (United States)

    Morales, Francisco J.; Reyes, Antonio; Cáceres, Noelia; Romero, Luis M.; Benitez, Francisco G.; Morgado, Joao; Duarte, Emanuel; Martins, Teresa

    2017-09-01

    A large percentage of transport infrastructures are composed of linear assets, such as roads and rail tracks. The large social and economic relevance of these constructions force the stakeholders to ensure a prolonged health/durability. Even though, inevitable malfunctioning, breaking down, and out-of-service periods arise randomly during the life cycle of the infrastructure. Predictive maintenance techniques tend to diminish the appearance of unpredicted failures and the execution of needed corrective interventions, envisaging the adequate interventions to be conducted before failures show up. This communication presents: i) A procedural approach, to be conducted, in order to collect the relevant information regarding the evolving state condition of the assets involved in all maintenance interventions; this reported and stored information constitutes a rich historical data base to train Machine Learning algorithms in order to generate reliable predictions of the interventions to be carried out in further time scenarios. ii) A schematic flow chart of the automatic learning procedure. iii) Self-learning rules from automatic learning from false positive/negatives. The description, testing, automatic learning approach and the outcomes of a pilot case are presented; finally some conclusions are outlined regarding the methodology proposed for improving the self-learning predictive capability.

  8. Resting heart rate variability predicts safety learning and fear extinction in an interoceptive fear conditioning paradigm.

    Directory of Open Access Journals (Sweden)

    Meike Pappens

    Full Text Available This study aimed to investigate whether interindividual differences in autonomic inhibitory control predict safety learning and fear extinction in an interoceptive fear conditioning paradigm. Data from a previously reported study (N = 40 were extended (N = 17 and re-analyzed to test whether healthy participants' resting heart rate variability (HRV - a proxy of cardiac vagal tone - predicts learning performance. The conditioned stimulus (CS was a slight sensation of breathlessness induced by a flow resistor, the unconditioned stimulus (US was an aversive short-lasting suffocation experience induced by a complete occlusion of the breathing circuitry. During acquisition, the paired group received 6 paired CS-US presentations; the control group received 6 explicitly unpaired CS-US presentations. In the extinction phase, both groups were exposed to 6 CS-only presentations. Measures included startle blink EMG, skin conductance responses (SCR and US-expectancy ratings. Resting HRV significantly predicted the startle blink EMG learning curves both during acquisition and extinction. In the unpaired group, higher levels of HRV at rest predicted safety learning to the CS during acquisition. In the paired group, higher levels of HRV were associated with better extinction. Our findings suggest that the strength or integrity of prefrontal inhibitory mechanisms involved in safety- and extinction learning can be indexed by HRV at rest.

  9. Empirical tests of the Gradual Learning Algorithm

    NARCIS (Netherlands)

    Boersma, P.; Hayes, B.

    1999-01-01

    The Gradual Learning Algorithm (Boersma 1997) is a constraint ranking algorithm for learning Optimality-theoretic grammars. The purpose of this article is to assess the capabilities of the Gradual Learning Algorithm, particularly in comparison with the Constraint Demotion algorithm of Tesar and

  10. Empirical tests of the Gradual Learning Algorithm

    NARCIS (Netherlands)

    Boersma, P.; Hayes, B.

    2001-01-01

    The Gradual Learning Algorithm (Boersma 1997) is a constraint-ranking algorithm for learning optimality-theoretic grammars. The purpose of this article is to assess the capabilities of the Gradual Learning Algorithm, particularly in comparison with the Constraint Demotion algorithm of Tesar and

  11. Functional connectivity between somatosensory and motor brain areas predicts individual differences in motor learning by observing.

    Science.gov (United States)

    McGregor, Heather R; Gribble, Paul L

    2017-08-01

    Action observation can facilitate the acquisition of novel motor skills; however, there is considerable individual variability in the extent to which observation promotes motor learning. Here we tested the hypothesis that individual differences in brain function or structure can predict subsequent observation-related gains in motor learning. Subjects underwent an anatomical MRI scan and resting-state fMRI scans to assess preobservation gray matter volume and preobservation resting-state functional connectivity (FC), respectively. On the following day, subjects observed a video of a tutor adapting her reaches to a novel force field. After observation, subjects performed reaches in a force field as a behavioral assessment of gains in motor learning resulting from observation. We found that individual differences in resting-state FC, but not gray matter volume, predicted postobservation gains in motor learning. Preobservation resting-state FC between left primary somatosensory cortex and bilateral dorsal premotor cortex, primary motor cortex, and primary somatosensory cortex and left superior parietal lobule was positively correlated with behavioral measures of postobservation motor learning. Sensory-motor resting-state FC can thus predict the extent to which observation will promote subsequent motor learning. NEW & NOTEWORTHY We show that individual differences in preobservation brain function can predict subsequent observation-related gains in motor learning. Preobservation resting-state functional connectivity within a sensory-motor network may be used as a biomarker for the extent to which observation promotes motor learning. This kind of information may be useful if observation is to be used as a way to boost neuroplasticity and sensory-motor recovery for patients undergoing rehabilitation for diseases that impair movement such as stroke. Copyright © 2017 the American Physiological Society.

  12. Machine learning approach for the outcome prediction of temporal lobe epilepsy surgery.

    Directory of Open Access Journals (Sweden)

    Rubén Armañanzas

    Full Text Available Epilepsy surgery is effective in reducing both the number and frequency of seizures, particularly in temporal lobe epilepsy (TLE. Nevertheless, a significant proportion of these patients continue suffering seizures after surgery. Here we used a machine learning approach to predict the outcome of epilepsy surgery based on supervised classification data mining taking into account not only the common clinical variables, but also pathological and neuropsychological evaluations. We have generated models capable of predicting whether a patient with TLE secondary to hippocampal sclerosis will fully recover from epilepsy or not. The machine learning analysis revealed that outcome could be predicted with an estimated accuracy of almost 90% using some clinical and neuropsychological features. Importantly, not all the features were needed to perform the prediction; some of them proved to be irrelevant to the prognosis. Personality style was found to be one of the key features to predict the outcome. Although we examined relatively few cases, findings were verified across all data, showing that the machine learning approach described in the present study may be a powerful method. Since neuropsychological assessment of epileptic patients is a standard protocol in the pre-surgical evaluation, we propose to include these specific psychological tests and machine learning tools to improve the selection of candidates for epilepsy surgery.

  13. Prediction of software operational reliability using testing environment factors

    International Nuclear Information System (INIS)

    Jung, Hoan Sung; Seong, Poong Hyun

    1995-01-01

    A number of software reliability models have been developed to estimate and to predict software reliability. However, there are no established standard models to quantify software reliability. Most models estimate the quality of software in reliability figures such as remaining faults, failure rate, or mean time to next failure at the testing phase, and they consider them ultimate indicators of software reliability. Experience shows that there is a large gap between predicted reliability during development and reliability measured during operation, which means that predicted reliability, or so-called test reliability, is not operational reliability. Customers prefer operational reliability to test reliability. In this study, we propose a method that predicts operational reliability rather than test reliability by introducing the testing environment factor that quantifies the changes in environments

  14. ERPs recorded during early second language exposure predict syntactic learning.

    Science.gov (United States)

    Batterink, Laura; Neville, Helen J

    2014-09-01

    Millions of adults worldwide are faced with the task of learning a second language (L2). Understanding the neural mechanisms that support this learning process is an important area of scientific inquiry. However, most previous studies on the neural mechanisms underlying L2 acquisition have focused on characterizing the results of learning, relying upon end-state outcome measures in which learning is assessed after it has occurred, rather than on the learning process itself. In this study, we adopted a novel and more direct approach to investigate neural mechanisms engaged during L2 learning, in which we recorded ERPs from beginning adult learners as they were exposed to an unfamiliar L2 for the first time. Learners' proficiency in the L2 was then assessed behaviorally using a grammaticality judgment task, and ERP data acquired during initial L2 exposure were sorted as a function of performance on this task. High-proficiency learners showed a larger N100 effect to open-class content words compared with closed-class function words, whereas low-proficiency learners did not show a significant N100 difference between open- and closed-class words. In contrast, amplitude of the N400 word category effect correlated with learners' L2 comprehension, rather than predicting syntactic learning. Taken together, these results indicate that learners who spontaneously direct greater attention to open- rather than closed-class words when processing L2 input show better syntactic learning, suggesting a link between selective attention to open-class content words and acquisition of basic morphosyntactic rules. These findings highlight the importance of selective attention mechanisms for L2 acquisition.

  15. Pipeline defect prediction using long range ultrasonic testing and intelligent processing

    International Nuclear Information System (INIS)

    Dino Isa; Rajprasad Rajkumar

    2009-01-01

    This paper deals with efforts to improve nondestructive testing (NDT) techniques by using artificial intelligence in detecting and predicting pipeline defects such as cracks and wall thinning. The main emphasis here will be on the prediction of corrosion type defects rather than just detection after the fact. Long range ultrasonic testing will be employed, where a ring of piezoelectric transducers are used to generate torsional guided waves. Various defects such as cracks as well as corrosion under insulation (CUI) will be simulated on a test pipe. The machine learning algorithm known as the Support Vector Machine (SVM) will be used to predict and classify transducer signals using regression and large margin classification. Regression results show that the SVM is able to accurately predict future defects based on trends of previous defect. The classification performance was also exceptional showing a facility to detect defects at different depths as well as for distinguishing closely spaced defects. (author)

  16. Machine learning application in online lending risk prediction

    OpenAIRE

    Yu, Xiaojiao

    2017-01-01

    Online leading has disrupted the traditional consumer banking sector with more effective loan processing. Risk prediction and monitoring is critical for the success of the business model. Traditional credit score models fall short in applying big data technology in building risk model. In this manuscript, data with various format and size were collected from public website, third-parties and assembled with client's loan application information data. Ensemble machine learning models, random fo...

  17. Pileup Subtraction and Jet Energy Prediction Using Machine Learning

    OpenAIRE

    Kong, Vein S; Li, Jiakun; Zhang, Yujia

    2015-01-01

    In the Large Hardron Collider (LHC), multiple proton-proton collisions cause pileup in reconstructing energy information for a single primary collision (jet). This project aims to select the most important features and create a model to accurately estimate jet energy. Different machine learning methods were explored, including linear regression, support vector regression and decision tree. The best result is obtained by linear regression with predictive features and the performance is improve...

  18. Do sophisticated epistemic beliefs predict meaningful learning? Findings from a structural equation model of undergraduate biology learning

    Science.gov (United States)

    Lee, Silvia Wen-Yu; Liang, Jyh-Chong; Tsai, Chin-Chung

    2016-10-01

    This study investigated the relationships among college students' epistemic beliefs in biology (EBB), conceptions of learning biology (COLB), and strategies of learning biology (SLB). EBB includes four dimensions, namely 'multiple-source,' 'uncertainty,' 'development,' and 'justification.' COLB is further divided into 'constructivist' and 'reproductive' conceptions, while SLB represents deep strategies and surface learning strategies. Questionnaire responses were gathered from 303 college students. The results of the confirmatory factor analysis and structural equation modelling showed acceptable model fits. Mediation testing further revealed two paths with complete mediation. In sum, students' epistemic beliefs of 'uncertainty' and 'justification' in biology were statistically significant in explaining the constructivist and reproductive COLB, respectively; and 'uncertainty' was statistically significant in explaining the deep SLB as well. The results of mediation testing further revealed that 'uncertainty' predicted surface strategies through the mediation of 'reproductive' conceptions; and the relationship between 'justification' and deep strategies was mediated by 'constructivist' COLB. This study provides evidence for the essential roles some epistemic beliefs play in predicting students' learning.

  19. Clinical chemistry in higher dimensions: Machine-learning and enhanced prediction from routine clinical chemistry data.

    Science.gov (United States)

    Richardson, Alice; Signor, Ben M; Lidbury, Brett A; Badrick, Tony

    2016-11-01

    Big Data is having an impact on many areas of research, not the least of which is biomedical science. In this review paper, big data and machine learning are defined in terms accessible to the clinical chemistry community. Seven myths associated with machine learning and big data are then presented, with the aim of managing expectation of machine learning amongst clinical chemists. The myths are illustrated with four examples investigating the relationship between biomarkers in liver function tests, enhanced laboratory prediction of hepatitis virus infection, the relationship between bilirubin and white cell count, and the relationship between red cell distribution width and laboratory prediction of anaemia. Copyright © 2016 The Canadian Society of Clinical Chemists. Published by Elsevier Inc. All rights reserved.

  20. What roles do errors serve in motor skill learning? An examination of two theoretical predictions.

    Science.gov (United States)

    Sanli, Elizabeth A; Lee, Timothy D

    2014-01-01

    Easy-to-difficult and difficult-to-easy progressions of task difficulty during skill acquisition were examined in 2 experiments that assessed retention, dual-task, and transfer tests of learning. Findings of the first experiment suggest that an easy-to difficult progression did not consistently induce implicit learning processes and was not consistently beneficial to performance under a secondary-task load. The findings of experiment two did not support the predictions made based on schema theory and only partially supported predictions based on reinvestment theory. The authors interpret these findings to suggest that the timing of error in relation to the difficulty of the task (functional task difficulty) plays a role in the transfer of learning to novel versions of a task.

  1. Machine learning derived risk prediction of anorexia nervosa.

    Science.gov (United States)

    Guo, Yiran; Wei, Zhi; Keating, Brendan J; Hakonarson, Hakon

    2016-01-20

    Anorexia nervosa (AN) is a complex psychiatric disease with a moderate to strong genetic contribution. In addition to conventional genome wide association (GWA) studies, researchers have been using machine learning methods in conjunction with genomic data to predict risk of diseases in which genetics play an important role. In this study, we collected whole genome genotyping data on 3940 AN cases and 9266 controls from the Genetic Consortium for Anorexia Nervosa (GCAN), the Wellcome Trust Case Control Consortium 3 (WTCCC3), Price Foundation Collaborative Group and the Children's Hospital of Philadelphia (CHOP), and applied machine learning methods for predicting AN disease risk. The prediction performance is measured by area under the receiver operating characteristic curve (AUC), indicating how well the model distinguishes cases from unaffected control subjects. Logistic regression model with the lasso penalty technique generated an AUC of 0.693, while Support Vector Machines and Gradient Boosted Trees reached AUC's of 0.691 and 0.623, respectively. Using different sample sizes, our results suggest that larger datasets are required to optimize the machine learning models and achieve higher AUC values. To our knowledge, this is the first attempt to assess AN risk based on genome wide genotype level data. Future integration of genomic, environmental and family-based information is likely to improve the AN risk evaluation process, eventually benefitting AN patients and families in the clinical setting.

  2. Using Deep Learning Model for Meteorological Satellite Cloud Image Prediction

    Science.gov (United States)

    Su, X.

    2017-12-01

    A satellite cloud image contains much weather information such as precipitation information. Short-time cloud movement forecast is important for precipitation forecast and is the primary means for typhoon monitoring. The traditional methods are mostly using the cloud feature matching and linear extrapolation to predict the cloud movement, which makes that the nonstationary process such as inversion and deformation during the movement of the cloud is basically not considered. It is still a hard task to predict cloud movement timely and correctly. As deep learning model could perform well in learning spatiotemporal features, to meet this challenge, we could regard cloud image prediction as a spatiotemporal sequence forecasting problem and introduce deep learning model to solve this problem. In this research, we use a variant of Gated-Recurrent-Unit(GRU) that has convolutional structures to deal with spatiotemporal features and build an end-to-end model to solve this forecast problem. In this model, both the input and output are spatiotemporal sequences. Compared to Convolutional LSTM(ConvLSTM) model, this model has lower amount of parameters. We imply this model on GOES satellite data and the model perform well.

  3. Human medial frontal cortex activity predicts learning from errors.

    Science.gov (United States)

    Hester, Robert; Barre, Natalie; Murphy, Kevin; Silk, Tim J; Mattingley, Jason B

    2008-08-01

    Learning from errors is a critical feature of human cognition. It underlies our ability to adapt to changing environmental demands and to tune behavior for optimal performance. The posterior medial frontal cortex (pMFC) has been implicated in the evaluation of errors to control behavior, although it has not previously been shown that activity in this region predicts learning from errors. Using functional magnetic resonance imaging, we examined activity in the pMFC during an associative learning task in which participants had to recall the spatial locations of 2-digit targets and were provided with immediate feedback regarding accuracy. Activity within the pMFC was significantly greater for errors that were subsequently corrected than for errors that were repeated. Moreover, pMFC activity during recall errors predicted future responses (correct vs. incorrect), despite a sizeable interval (on average 70 s) between an error and the next presentation of the same recall probe. Activity within the hippocampus also predicted future performance and correlated with error-feedback-related pMFC activity. A relationship between performance expectations and pMFC activity, in the absence of differing reinforcement value for errors, is consistent with the idea that error-related pMFC activity reflects the extent to which an outcome is "worse than expected."

  4. Sequence-based prediction of protein protein interaction using a deep-learning algorithm.

    Science.gov (United States)

    Sun, Tanlin; Zhou, Bo; Lai, Luhua; Pei, Jianfeng

    2017-05-25

    Protein-protein interactions (PPIs) are critical for many biological processes. It is therefore important to develop accurate high-throughput methods for identifying PPI to better understand protein function, disease occurrence, and therapy design. Though various computational methods for predicting PPI have been developed, their robustness for prediction with external datasets is unknown. Deep-learning algorithms have achieved successful results in diverse areas, but their effectiveness for PPI prediction has not been tested. We used a stacked autoencoder, a type of deep-learning algorithm, to study the sequence-based PPI prediction. The best model achieved an average accuracy of 97.19% with 10-fold cross-validation. The prediction accuracies for various external datasets ranged from 87.99% to 99.21%, which are superior to those achieved with previous methods. To our knowledge, this research is the first to apply a deep-learning algorithm to sequence-based PPI prediction, and the results demonstrate its potential in this field.

  5. Wind Power Ramp Events Prediction with Hybrid Machine Learning Regression Techniques and Reanalysis Data

    Directory of Open Access Journals (Sweden)

    Laura Cornejo-Bueno

    2017-11-01

    Full Text Available Wind Power Ramp Events (WPREs are large fluctuations of wind power in a short time interval, which lead to strong, undesirable variations in the electric power produced by a wind farm. Its accurate prediction is important in the effort of efficiently integrating wind energy in the electric system, without affecting considerably its stability, robustness and resilience. In this paper, we tackle the problem of predicting WPREs by applying Machine Learning (ML regression techniques. Our approach consists of using variables from atmospheric reanalysis data as predictive inputs for the learning machine, which opens the possibility of hybridizing numerical-physical weather models with ML techniques for WPREs prediction in real systems. Specifically, we have explored the feasibility of a number of state-of-the-art ML regression techniques, such as support vector regression, artificial neural networks (multi-layer perceptrons and extreme learning machines and Gaussian processes to solve the problem. Furthermore, the ERA-Interim reanalysis from the European Center for Medium-Range Weather Forecasts is the one used in this paper because of its accuracy and high resolution (in both spatial and temporal domains. Aiming at validating the feasibility of our predicting approach, we have carried out an extensive experimental work using real data from three wind farms in Spain, discussing the performance of the different ML regression tested in this wind power ramp event prediction problem.

  6. Sleep Quality Prediction From Wearable Data Using Deep Learning.

    Science.gov (United States)

    Sathyanarayana, Aarti; Joty, Shafiq; Fernandez-Luque, Luis; Ofli, Ferda; Srivastava, Jaideep; Elmagarmid, Ahmed; Arora, Teresa; Taheri, Shahrad

    2016-11-04

    The importance of sleep is paramount to health. Insufficient sleep can reduce physical, emotional, and mental well-being and can lead to a multitude of health complications among people with chronic conditions. Physical activity and sleep are highly interrelated health behaviors. Our physical activity during the day (ie, awake time) influences our quality of sleep, and vice versa. The current popularity of wearables for tracking physical activity and sleep, including actigraphy devices, can foster the development of new advanced data analytics. This can help to develop new electronic health (eHealth) applications and provide more insights into sleep science. The objective of this study was to evaluate the feasibility of predicting sleep quality (ie, poor or adequate sleep efficiency) given the physical activity wearable data during awake time. In this study, we focused on predicting good or poor sleep efficiency as an indicator of sleep quality. Actigraphy sensors are wearable medical devices used to study sleep and physical activity patterns. The dataset used in our experiments contained the complete actigraphy data from a subset of 92 adolescents over 1 full week. Physical activity data during awake time was used to create predictive models for sleep quality, in particular, poor or good sleep efficiency. The physical activity data from sleep time was used for the evaluation. We compared the predictive performance of traditional logistic regression with more advanced deep learning methods: multilayer perceptron (MLP), convolutional neural network (CNN), simple Elman-type recurrent neural network (RNN), long short-term memory (LSTM-RNN), and a time-batched version of LSTM-RNN (TB-LSTM). Deep learning models were able to predict the quality of sleep (ie, poor or good sleep efficiency) based on wearable data from awake periods. More specifically, the deep learning methods performed better than traditional logistic regression. “CNN had the highest specificity and

  7. Test-Enhanced Learning in an Immunology and Infectious Disease Medicinal Chemistry/Pharmacology Course.

    Science.gov (United States)

    Hernick, Marcy

    2015-09-25

    Objective. To develop a series of active-learning modules that would improve pharmacy students' performance on summative assessments. Design. A series of optional online active-learning modules containing questions with multiple formats for topics in a first-year (P1) course was created using a test-enhanced learning approach. A subset of module questions was modified and included on summative assessments. Assessment. Student performance on module questions improved with repeated attempts and was predictive of student performance on summative assessments. Performance on examination questions was higher for students with access to modules than for those without access to modules. Module use appeared to have the most impact on low performing students. Conclusion. Test-enhanced learning modules with immediate feedback provide pharmacy students with a learning tool that improves student performance on summative assessments and also may improve metacognitive and test-taking skills.

  8. Evaluating the Predictive Validity of Graduate Management Admission Test Scores

    Science.gov (United States)

    Sireci, Stephen G.; Talento-Miller, Eileen

    2006-01-01

    Admissions data and first-year grade point average (GPA) data from 11 graduate management schools were analyzed to evaluate the predictive validity of Graduate Management Admission Test[R] (GMAT[R]) scores and the extent to which predictive validity held across sex and race/ethnicity. The results indicated GMAT verbal and quantitative scores had…

  9. Machine learning models in breast cancer survival prediction.

    Science.gov (United States)

    Montazeri, Mitra; Montazeri, Mohadeseh; Montazeri, Mahdieh; Beigzadeh, Amin

    2016-01-01

    Breast cancer is one of the most common cancers with a high mortality rate among women. With the early diagnosis of breast cancer survival will increase from 56% to more than 86%. Therefore, an accurate and reliable system is necessary for the early diagnosis of this cancer. The proposed model is the combination of rules and different machine learning techniques. Machine learning models can help physicians to reduce the number of false decisions. They try to exploit patterns and relationships among a large number of cases and predict the outcome of a disease using historical cases stored in datasets. The objective of this study is to propose a rule-based classification method with machine learning techniques for the prediction of different types of Breast cancer survival. We use a dataset with eight attributes that include the records of 900 patients in which 876 patients (97.3%) and 24 (2.7%) patients were females and males respectively. Naive Bayes (NB), Trees Random Forest (TRF), 1-Nearest Neighbor (1NN), AdaBoost (AD), Support Vector Machine (SVM), RBF Network (RBFN), and Multilayer Perceptron (MLP) machine learning techniques with 10-cross fold technique were used with the proposed model for the prediction of breast cancer survival. The performance of machine learning techniques were evaluated with accuracy, precision, sensitivity, specificity, and area under ROC curve. Out of 900 patients, 803 patients and 97 patients were alive and dead, respectively. In this study, Trees Random Forest (TRF) technique showed better results in comparison to other techniques (NB, 1NN, AD, SVM and RBFN, MLP). The accuracy, sensitivity and the area under ROC curve of TRF are 96%, 96%, 93%, respectively. However, 1NN machine learning technique provided poor performance (accuracy 91%, sensitivity 91% and area under ROC curve 78%). This study demonstrates that Trees Random Forest model (TRF) which is a rule-based classification model was the best model with the highest level of

  10. Can personality predict individual differences in brook trout spatial learning ability?

    Science.gov (United States)

    White, S L; Wagner, T; Gowan, C; Braithwaite, V A

    2017-08-01

    While differences in individual personality are common in animal populations, understanding the ecological significance of variation has not yet been resolved. Evidence suggests that personality may influence learning and memory; a finding that could improve our understanding of the evolutionary processes that produce and maintain intraspecific behavioural heterogeneity. Here, we tested whether boldness, the most studied personality trait in fish, could predict learning ability in brook trout. After quantifying boldness, fish were trained to find a hidden food patch in a maze environment. Stable landmark cues were provided to indicate the location of food and, at the conclusion of training, cues were rearranged to test for learning. There was a negative relationship between boldness and learning as shy fish were increasingly more successful at navigating the maze and locating food during training trials compared to bold fish. In the altered testing environment, only shy fish continued using cues to search for food. Overall, the learning rate of bold fish was found to be lower than that of shy fish for several metrics suggesting that personality could have widespread effects on behaviour. Because learning can increase plasticity to environmental change, these results have significant implications for fish conservation. Copyright © 2016 Elsevier B.V. All rights reserved.

  11. Can personality predict individual differences in brook trout spatial learning ability?

    Science.gov (United States)

    White, S.L.; Wagner, Tyler; Gowan, C.; Braithwaite, V.A.

    2017-01-01

    While differences in individual personality are common in animal populations, understanding the ecological significance of variation has not yet been resolved. Evidence suggests that personality may influence learning and memory; a finding that could improve our understanding of the evolutionary processes that produce and maintain intraspecific behavioural heterogeneity. Here, we tested whether boldness, the most studied personality trait in fish, could predict learning ability in brook trout. After quantifying boldness, fish were trained to find a hidden food patch in a maze environment. Stable landmark cues were provided to indicate the location of food and, at the conclusion of training, cues were rearranged to test for learning. There was a negative relationship between boldness and learning as shy fish were increasingly more successful at navigating the maze and locating food during training trials compared to bold fish. In the altered testing environment, only shy fish continued using cues to search for food. Overall, the learning rate of bold fish was found to be lower than that of shy fish for several metrics suggesting that personality could have widespread effects on behaviour. Because learning can increase plasticity to environmental change, these results have significant implications for fish conservation.

  12. Machine Learning Techniques for Prediction of Early Childhood Obesity.

    Science.gov (United States)

    Dugan, T M; Mukhopadhyay, S; Carroll, A; Downs, S

    2015-01-01

    This paper aims to predict childhood obesity after age two, using only data collected prior to the second birthday by a clinical decision support system called CHICA. Analyses of six different machine learning methods: RandomTree, RandomForest, J48, ID3, Naïve Bayes, and Bayes trained on CHICA data show that an accurate, sensitive model can be created. Of the methods analyzed, the ID3 model trained on the CHICA dataset proved the best overall performance with accuracy of 85% and sensitivity of 89%. Additionally, the ID3 model had a positive predictive value of 84% and a negative predictive value of 88%. The structure of the tree also gives insight into the strongest predictors of future obesity in children. Many of the strongest predictors seen in the ID3 modeling of the CHICA dataset have been independently validated in the literature as correlated with obesity, thereby supporting the validity of the model. This study demonstrated that data from a production clinical decision support system can be used to build an accurate machine learning model to predict obesity in children after age two.

  13. Machine learning applied to the prediction of citrus production

    Energy Technology Data Exchange (ETDEWEB)

    Díaz, I.; Mazza, S.M.; Combarro, E.F.; Giménez, L.I.; Gaiad, J.E.

    2017-07-01

    An in-depth knowledge about variables affecting production is required in order to predict global production and take decisions in agriculture. Machine learning is a technique used in agricultural planning and precision agriculture. This work (i) studies the effectiveness of machine learning techniques for predicting orchards production; and (ii) variables affecting this production were also identified. Data from 964 orchards of lemon, mandarin, and orange in Corrientes, Argentina are analysed. Graphic and analytical descriptive statistics, correlation coefficients, principal component analysis and Biplot were performed. Production was predicted via M5-Prime, a model regression tree constructor which produces a classification based on piecewise linear functions. For all the species studied, the most informative variable was the trees' age; in mandarin and orange orchards, age was followed by between and within row distances; irrigation also affected mandarin production. Also, the performance of M5-Prime in the prediction of production is adequate, as shown when measured with correlation coefficients (~0.8) and relative mean absolute error (~0.1). These results show that M5-Prime is an appropriate method to classify citrus orchards according to production and, in addition, it allows for identifying the most informative variables affecting production by tree.

  14. Machine learning applied to the prediction of citrus production

    International Nuclear Information System (INIS)

    Díaz, I.; Mazza, S.M.; Combarro, E.F.; Giménez, L.I.; Gaiad, J.E.

    2017-01-01

    An in-depth knowledge about variables affecting production is required in order to predict global production and take decisions in agriculture. Machine learning is a technique used in agricultural planning and precision agriculture. This work (i) studies the effectiveness of machine learning techniques for predicting orchards production; and (ii) variables affecting this production were also identified. Data from 964 orchards of lemon, mandarin, and orange in Corrientes, Argentina are analysed. Graphic and analytical descriptive statistics, correlation coefficients, principal component analysis and Biplot were performed. Production was predicted via M5-Prime, a model regression tree constructor which produces a classification based on piecewise linear functions. For all the species studied, the most informative variable was the trees' age; in mandarin and orange orchards, age was followed by between and within row distances; irrigation also affected mandarin production. Also, the performance of M5-Prime in the prediction of production is adequate, as shown when measured with correlation coefficients (~0.8) and relative mean absolute error (~0.1). These results show that M5-Prime is an appropriate method to classify citrus orchards according to production and, in addition, it allows for identifying the most informative variables affecting production by tree.

  15. Machine learning applied to the prediction of citrus production

    Directory of Open Access Journals (Sweden)

    Irene Díaz

    2017-07-01

    Full Text Available An in-depth knowledge about variables affecting production is required in order to predict global production and take decisions in agriculture. Machine learning is a technique used in agricultural planning and precision agriculture. This work (i studies the effectiveness of machine learning techniques for predicting orchards production; and (ii variables affecting this production were also identified. Data from 964 orchards of lemon, mandarin, and orange in Corrientes, Argentina are analysed. Graphic and analytical descriptive statistics, correlation coefficients, principal component analysis and Biplot were performed. Production was predicted via M5-Prime, a model regression tree constructor which produces a classification based on piecewise linear functions. For all the species studied, the most informative variable was the trees’ age; in mandarin and orange orchards, age was followed by between and within row distances; irrigation also affected mandarin production. Also, the performance of M5-Prime in the prediction of production is adequate, as shown when measured with correlation coefficients (~0.8 and relative mean absolute error (~0.1. These results show that M5-Prime is an appropriate method to classify citrus orchards according to production and, in addition, it allows for identifying the most informative variables affecting production by tree.

  16. When bad stress goes good: increased threat reactivity predicts improved category learning performance.

    Science.gov (United States)

    Ell, Shawn W; Cosley, Brandon; McCoy, Shannon K

    2011-02-01

    The way in which we respond to everyday stressors can have a profound impact on cognitive functioning. Maladaptive stress responses in particular are generally associated with impaired cognitive performance. We argue, however, that the cognitive system mediating task performance is also a critical determinant of the stress-cognition relationship. Consistent with this prediction, we observed that stress reactivity consistent with a maladaptive, threat response differentially predicted performance on two categorization tasks. Increased threat reactivity predicted enhanced performance on an information-integration task (i.e., learning is thought to depend upon a procedural-based memory system), and a (nonsignificant) trend for impaired performance on a rule-based task (i.e., learning is thought to depend upon a hypothesis-testing system). These data suggest that it is critical to consider both variability in the stress response and variability in the cognitive system mediating task performance in order to fully understand the stress-cognition relationship.

  17. Prediction of Student Dropout in E-Learning Program Through the Use of Machine Learning Method

    OpenAIRE

    Mingjie Tan; Peiji Shao

    2015-01-01

    The high rate of dropout is a serious problem in E-learning program. Thus it has received extensive concern from the education administrators and researchers. Predicting the potential dropout students is a workable solution to prevent dropout. Based on the analysis of related literature, this study selected student’s personal characteristic and academic performance as input attributions. Prediction models were developed using Artificial Neural Network (ANN), Decision Tree (DT) and Bayesian Ne...

  18. Prospective detection of large prediction errors: a hypothesis testing approach

    International Nuclear Information System (INIS)

    Ruan, Dan

    2010-01-01

    Real-time motion management is important in radiotherapy. In addition to effective monitoring schemes, prediction is required to compensate for system latency, so that treatment can be synchronized with tumor motion. However, it is difficult to predict tumor motion at all times, and it is critical to determine when large prediction errors may occur. Such information can be used to pause the treatment beam or adjust monitoring/prediction schemes. In this study, we propose a hypothesis testing approach for detecting instants corresponding to potentially large prediction errors in real time. We treat the future tumor location as a random variable, and obtain its empirical probability distribution with the kernel density estimation-based method. Under the null hypothesis, the model probability is assumed to be a concentrated Gaussian centered at the prediction output. Under the alternative hypothesis, the model distribution is assumed to be non-informative uniform, which reflects the situation that the future position cannot be inferred reliably. We derive the likelihood ratio test (LRT) for this hypothesis testing problem and show that with the method of moments for estimating the null hypothesis Gaussian parameters, the LRT reduces to a simple test on the empirical variance of the predictive random variable. This conforms to the intuition to expect a (potentially) large prediction error when the estimate is associated with high uncertainty, and to expect an accurate prediction when the uncertainty level is low. We tested the proposed method on patient-derived respiratory traces. The 'ground-truth' prediction error was evaluated by comparing the prediction values with retrospective observations, and the large prediction regions were subsequently delineated by thresholding the prediction errors. The receiver operating characteristic curve was used to describe the performance of the proposed hypothesis testing method. Clinical implication was represented by miss

  19. ComplexContact: a web server for inter-protein contact prediction using deep learning

    KAUST Repository

    Zeng, Hong; Wang, Sheng; Zhou, Tianming; Zhao, Feifeng; Li, Xiufeng; Wu, Qing; Xu, Jinbo

    2018-01-01

    ComplexContact (http://raptorx2.uchicago.edu/ComplexContact/) is a web server for sequence-based interfacial residue-residue contact prediction of a putative protein complex. Interfacial residue-residue contacts are critical for understanding how proteins form complex and interact at residue level. When receiving a pair of protein sequences, ComplexContact first searches for their sequence homologs and builds two paired multiple sequence alignments (MSA), then it applies co-evolution analysis and a CASP-winning deep learning (DL) method to predict interfacial contacts from paired MSAs and visualizes the prediction as an image. The DL method was originally developed for intra-protein contact prediction and performed the best in CASP12. Our large-scale experimental test further shows that ComplexContact greatly outperforms pure co-evolution methods for inter-protein contact prediction, regardless of the species.

  20. ComplexContact: a web server for inter-protein contact prediction using deep learning

    KAUST Repository

    Zeng, Hong

    2018-05-20

    ComplexContact (http://raptorx2.uchicago.edu/ComplexContact/) is a web server for sequence-based interfacial residue-residue contact prediction of a putative protein complex. Interfacial residue-residue contacts are critical for understanding how proteins form complex and interact at residue level. When receiving a pair of protein sequences, ComplexContact first searches for their sequence homologs and builds two paired multiple sequence alignments (MSA), then it applies co-evolution analysis and a CASP-winning deep learning (DL) method to predict interfacial contacts from paired MSAs and visualizes the prediction as an image. The DL method was originally developed for intra-protein contact prediction and performed the best in CASP12. Our large-scale experimental test further shows that ComplexContact greatly outperforms pure co-evolution methods for inter-protein contact prediction, regardless of the species.

  1. ComplexContact: a web server for inter-protein contact prediction using deep learning.

    Science.gov (United States)

    Zeng, Hong; Wang, Sheng; Zhou, Tianming; Zhao, Feifeng; Li, Xiufeng; Wu, Qing; Xu, Jinbo

    2018-05-22

    ComplexContact (http://raptorx2.uchicago.edu/ComplexContact/) is a web server for sequence-based interfacial residue-residue contact prediction of a putative protein complex. Interfacial residue-residue contacts are critical for understanding how proteins form complex and interact at residue level. When receiving a pair of protein sequences, ComplexContact first searches for their sequence homologs and builds two paired multiple sequence alignments (MSA), then it applies co-evolution analysis and a CASP-winning deep learning (DL) method to predict interfacial contacts from paired MSAs and visualizes the prediction as an image. The DL method was originally developed for intra-protein contact prediction and performed the best in CASP12. Our large-scale experimental test further shows that ComplexContact greatly outperforms pure co-evolution methods for inter-protein contact prediction, regardless of the species.

  2. PEDLA: predicting enhancers with a deep learning-based algorithmic framework.

    Science.gov (United States)

    Liu, Feng; Li, Hao; Ren, Chao; Bo, Xiaochen; Shu, Wenjie

    2016-06-22

    Transcriptional enhancers are non-coding segments of DNA that play a central role in the spatiotemporal regulation of gene expression programs. However, systematically and precisely predicting enhancers remain a major challenge. Although existing methods have achieved some success in enhancer prediction, they still suffer from many issues. We developed a deep learning-based algorithmic framework named PEDLA (https://github.com/wenjiegroup/PEDLA), which can directly learn an enhancer predictor from massively heterogeneous data and generalize in ways that are mostly consistent across various cell types/tissues. We first trained PEDLA with 1,114-dimensional heterogeneous features in H1 cells, and demonstrated that PEDLA framework integrates diverse heterogeneous features and gives state-of-the-art performance relative to five existing methods for enhancer prediction. We further extended PEDLA to iteratively learn from 22 training cell types/tissues. Our results showed that PEDLA manifested superior performance consistency in both training and independent test sets. On average, PEDLA achieved 95.0% accuracy and a 96.8% geometric mean (GM) of sensitivity and specificity across 22 training cell types/tissues, as well as 95.7% accuracy and a 96.8% GM across 20 independent test cell types/tissues. Together, our work illustrates the power of harnessing state-of-the-art deep learning techniques to consistently identify regulatory elements at a genome-wide scale from massively heterogeneous data across diverse cell types/tissues.

  3. Testing the predictive power of nuclear mass models

    International Nuclear Information System (INIS)

    Mendoza-Temis, J.; Morales, I.; Barea, J.; Frank, A.; Hirsch, J.G.; Vieyra, J.C. Lopez; Van Isacker, P.; Velazquez, V.

    2008-01-01

    A number of tests are introduced which probe the ability of nuclear mass models to extrapolate. Three models are analyzed in detail: the liquid drop model, the liquid drop model plus empirical shell corrections and the Duflo-Zuker mass formula. If predicted nuclei are close to the fitted ones, average errors in predicted and fitted masses are similar. However, the challenge of predicting nuclear masses in a region stabilized by shell effects (e.g., the lead region) is far more difficult. The Duflo-Zuker mass formula emerges as a powerful predictive tool

  4. Machine Learning and Neurosurgical Outcome Prediction: A Systematic Review.

    Science.gov (United States)

    Senders, Joeky T; Staples, Patrick C; Karhade, Aditya V; Zaki, Mark M; Gormley, William B; Broekman, Marike L D; Smith, Timothy R; Arnaout, Omar

    2018-01-01

    Accurate measurement of surgical outcomes is highly desirable to optimize surgical decision-making. An important element of surgical decision making is identification of the patient cohort that will benefit from surgery before the intervention. Machine learning (ML) enables computers to learn from previous data to make accurate predictions on new data. In this systematic review, we evaluate the potential of ML for neurosurgical outcome prediction. A systematic search in the PubMed and Embase databases was performed to identify all potential relevant studies up to January 1, 2017. Thirty studies were identified that evaluated ML algorithms used as prediction models for survival, recurrence, symptom improvement, and adverse events in patients undergoing surgery for epilepsy, brain tumor, spinal lesions, neurovascular disease, movement disorders, traumatic brain injury, and hydrocephalus. Depending on the specific prediction task evaluated and the type of input features included, ML models predicted outcomes after neurosurgery with a median accuracy and area under the receiver operating curve of 94.5% and 0.83, respectively. Compared with logistic regression, ML models performed significantly better and showed a median absolute improvement in accuracy and area under the receiver operating curve of 15% and 0.06, respectively. Some studies also demonstrated a better performance in ML models compared with established prognostic indices and clinical experts. In the research setting, ML has been studied extensively, demonstrating an excellent performance in outcome prediction for a wide range of neurosurgical conditions. However, future studies should investigate how ML can be implemented as a practical tool supporting neurosurgical care. Copyright © 2017 Elsevier Inc. All rights reserved.

  5. Episodic Memory Encoding Interferes with Reward Learning and Decreases Striatal Prediction Errors

    Science.gov (United States)

    Braun, Erin Kendall; Daw, Nathaniel D.

    2014-01-01

    Learning is essential for adaptive decision making. The striatum and its dopaminergic inputs are known to support incremental reward-based learning, while the hippocampus is known to support encoding of single events (episodic memory). Although traditionally studied separately, in even simple experiences, these two types of learning are likely to co-occur and may interact. Here we sought to understand the nature of this interaction by examining how incremental reward learning is related to concurrent episodic memory encoding. During the experiment, human participants made choices between two options (colored squares), each associated with a drifting probability of reward, with the goal of earning as much money as possible. Incidental, trial-unique object pictures, unrelated to the choice, were overlaid on each option. The next day, participants were given a surprise memory test for these pictures. We found that better episodic memory was related to a decreased influence of recent reward experience on choice, both within and across participants. fMRI analyses further revealed that during learning the canonical striatal reward prediction error signal was significantly weaker when episodic memory was stronger. This decrease in reward prediction error signals in the striatum was associated with enhanced functional connectivity between the hippocampus and striatum at the time of choice. Our results suggest a mechanism by which memory encoding may compete for striatal processing and provide insight into how interactions between different forms of learning guide reward-based decision making. PMID:25378157

  6. Episodic memory encoding interferes with reward learning and decreases striatal prediction errors.

    Science.gov (United States)

    Wimmer, G Elliott; Braun, Erin Kendall; Daw, Nathaniel D; Shohamy, Daphna

    2014-11-05

    Learning is essential for adaptive decision making. The striatum and its dopaminergic inputs are known to support incremental reward-based learning, while the hippocampus is known to support encoding of single events (episodic memory). Although traditionally studied separately, in even simple experiences, these two types of learning are likely to co-occur and may interact. Here we sought to understand the nature of this interaction by examining how incremental reward learning is related to concurrent episodic memory encoding. During the experiment, human participants made choices between two options (colored squares), each associated with a drifting probability of reward, with the goal of earning as much money as possible. Incidental, trial-unique object pictures, unrelated to the choice, were overlaid on each option. The next day, participants were given a surprise memory test for these pictures. We found that better episodic memory was related to a decreased influence of recent reward experience on choice, both within and across participants. fMRI analyses further revealed that during learning the canonical striatal reward prediction error signal was significantly weaker when episodic memory was stronger. This decrease in reward prediction error signals in the striatum was associated with enhanced functional connectivity between the hippocampus and striatum at the time of choice. Our results suggest a mechanism by which memory encoding may compete for striatal processing and provide insight into how interactions between different forms of learning guide reward-based decision making. Copyright © 2014 the authors 0270-6474/14/3414901-12$15.00/0.

  7. Optimizing Multiple-Choice Tests as Learning Events

    Science.gov (United States)

    Little, Jeri Lynn

    2011-01-01

    Although generally used for assessment, tests can also serve as tools for learning--but different test formats may not be equally beneficial. Specifically, research has shown multiple-choice tests to be less effective than cued-recall tests in improving the later retention of the tested information (e.g., see meta-analysis by Hamaker, 1986),…

  8. The Relationship between Learning Style, Test Anxiety and Academic Achievement

    Science.gov (United States)

    Yazici, Kubilay

    2017-01-01

    This study aimed to investigate the relationship between social studies pre-service teachers' (SSPTs) learning style, test anxiety and academic achievement. A total of 315 SSPTs participated in the study. Data were collected using Turkish versions of Grasha-Reichmann learning style scale (GRLSS) and test anxiety scale (TAS) by Spielberger.…

  9. Dopamine prediction errors in reward learning and addiction: from theory to neural circuitry

    Science.gov (United States)

    Keiflin, Ronald; Janak, Patricia H.

    2015-01-01

    Summary Midbrain dopamine (DA) neurons are proposed to signal reward prediction error (RPE), a fundamental parameter in associative learning models. This RPE hypothesis provides a compelling theoretical framework for understanding DA function in reward learning and addiction. New studies support a causal role for DA-mediated RPE activity in promoting learning about natural reward; however, this question has not been explicitly tested in the context of drug addiction. In this review, we integrate theoretical models with experimental findings on the activity of DA systems, and on the causal role of specific neuronal projections and cell types, to provide a circuit-based framework for probing DA-RPE function in addiction. By examining error-encoding DA neurons in the neural network in which they are embedded, hypotheses regarding circuit-level adaptations that possibly contribute to pathological error-signaling and addiction can be formulated and tested. PMID:26494275

  10. Dopamine Prediction Errors in Reward Learning and Addiction: From Theory to Neural Circuitry.

    Science.gov (United States)

    Keiflin, Ronald; Janak, Patricia H

    2015-10-21

    Midbrain dopamine (DA) neurons are proposed to signal reward prediction error (RPE), a fundamental parameter in associative learning models. This RPE hypothesis provides a compelling theoretical framework for understanding DA function in reward learning and addiction. New studies support a causal role for DA-mediated RPE activity in promoting learning about natural reward; however, this question has not been explicitly tested in the context of drug addiction. In this review, we integrate theoretical models with experimental findings on the activity of DA systems, and on the causal role of specific neuronal projections and cell types, to provide a circuit-based framework for probing DA-RPE function in addiction. By examining error-encoding DA neurons in the neural network in which they are embedded, hypotheses regarding circuit-level adaptations that possibly contribute to pathological error signaling and addiction can be formulated and tested. Copyright © 2015 Elsevier Inc. All rights reserved.

  11. PBF/LOFT Lead Rod Test Program experiment predictions document

    International Nuclear Information System (INIS)

    Varacalle, D.J.; Cox, W.R.; Niebruegge, D.A.; Seiber, S.J.; Brake, T.E.; Driskell, W.E.; Nigg, D.W.; Tolman, E.L.

    1978-12-01

    The PBF/LOFT Lead Rod (LLR) Test Program is being conducted to provide experimental information on the behavior of nuclear fuel under normal and accident conditions in the Power Burst Facility (PBF) at the Idaho National Engineering Laboratory. The PBF/LLR tests are designed to simulate the test conditions for the LOFT Power Ascension Tests L2-3 through L2-5. The test program has been designed to provide a parametric evaluation of the LOFT fuel (center and peripheral modules) over a wide range of power. This report presents the experiment predictions for the three four-rod LOCA tests

  12. Testing protects against proactive interference in face-name learning.

    Science.gov (United States)

    Weinstein, Yana; McDermott, Kathleen B; Szpunar, Karl K

    2011-06-01

    Learning face-name pairings at a social function becomes increasingly more difficult the more individuals one meets. This phenomenon is attributable to proactive interference--the negative influence of prior learning on subsequent learning. Recent evidence suggests that taking a memory test can alleviate proactive interference in verbal list learning paradigms. We apply this technique to face-name pair learning. Participants studied four lists of 12 face-name pairings and either attempted to name the 12 faces just studied after every list or did not. Recall attempts after every list improved learning of the fourth list by over 100%. Moreover, no reduction in learning of face-name pairings occurred from list 1 to list 4 for participants who attempted to name studied faces between lists. These results suggest that testing oneself on the names of a group of new acquaintances before moving on to the next group is an effective mnemonic technique for social functions.

  13. Evaluation of Deep Learning Models for Predicting CO2 Flux

    Science.gov (United States)

    Halem, M.; Nguyen, P.; Frankel, D.

    2017-12-01

    Artificial neural networks have been employed to calculate surface flux measurements from station data because they are able to fit highly nonlinear relations between input and output variables without knowing the detail relationships between the variables. However, the accuracy in performing neural net estimates of CO2 flux from observations of CO2 and other atmospheric variables is influenced by the architecture of the neural model, the availability, and complexity of interactions between physical variables such as wind, temperature, and indirect variables like latent heat, and sensible heat, etc. We evaluate two deep learning models, feed forward and recurrent neural network models to learn how they each respond to the physical measurements, time dependency of the measurements of CO2 concentration, humidity, pressure, temperature, wind speed etc. for predicting the CO2 flux. In this paper, we focus on a) building neural network models for estimating CO2 flux based on DOE data from tower Atmospheric Radiation Measurement data; b) evaluating the impact of choosing the surface variables and model hyper-parameters on the accuracy and predictions of surface flux; c) assessing the applicability of the neural network models on estimate CO2 flux by using OCO-2 satellite data; d) studying the efficiency of using GPU-acceleration for neural network performance using IBM Power AI deep learning software and packages on IBM Minsky system.

  14. Machine learning for outcome prediction of acute ischemic stroke post intra-arterial therapy.

    Directory of Open Access Journals (Sweden)

    Hamed Asadi

    Full Text Available INTRODUCTION: Stroke is a major cause of death and disability. Accurately predicting stroke outcome from a set of predictive variables may identify high-risk patients and guide treatment approaches, leading to decreased morbidity. Logistic regression models allow for the identification and validation of predictive variables. However, advanced machine learning algorithms offer an alternative, in particular, for large-scale multi-institutional data, with the advantage of easily incorporating newly available data to improve prediction performance. Our aim was to design and compare different machine learning methods, capable of predicting the outcome of endovascular intervention in acute anterior circulation ischaemic stroke. METHOD: We conducted a retrospective study of a prospectively collected database of acute ischaemic stroke treated by endovascular intervention. Using SPSS®, MATLAB®, and Rapidminer®, classical statistics as well as artificial neural network and support vector algorithms were applied to design a supervised machine capable of classifying these predictors into potential good and poor outcomes. These algorithms were trained, validated and tested using randomly divided data. RESULTS: We included 107 consecutive acute anterior circulation ischaemic stroke patients treated by endovascular technique. Sixty-six were male and the mean age of 65.3. All the available demographic, procedural and clinical factors were included into the models. The final confusion matrix of the neural network, demonstrated an overall congruency of ∼ 80% between the target and output classes, with favourable receiving operative characteristics. However, after optimisation, the support vector machine had a relatively better performance, with a root mean squared error of 2.064 (SD: ± 0.408. DISCUSSION: We showed promising accuracy of outcome prediction, using supervised machine learning algorithms, with potential for incorporation of larger multicenter

  15. Reinforcement Learning Based Novel Adaptive Learning Framework for Smart Grid Prediction

    Directory of Open Access Journals (Sweden)

    Tian Li

    2017-01-01

    Full Text Available Smart grid is a potential infrastructure to supply electricity demand for end users in a safe and reliable manner. With the rapid increase of the share of renewable energy and controllable loads in smart grid, the operation uncertainty of smart grid has increased briskly during recent years. The forecast is responsible for the safety and economic operation of the smart grid. However, most existing forecast methods cannot account for the smart grid due to the disabilities to adapt to the varying operational conditions. In this paper, reinforcement learning is firstly exploited to develop an online learning framework for the smart grid. With the capability of multitime scale resolution, wavelet neural network has been adopted in the online learning framework to yield reinforcement learning and wavelet neural network (RLWNN based adaptive learning scheme. The simulations on two typical prediction problems in smart grid, including wind power prediction and load forecast, validate the effectiveness and the scalability of the proposed RLWNN based learning framework and algorithm.

  16. An Approach for Predicting Essential Genes Using Multiple Homology Mapping and Machine Learning Algorithms.

    Science.gov (United States)

    Hua, Hong-Li; Zhang, Fa-Zhan; Labena, Abraham Alemayehu; Dong, Chuan; Jin, Yan-Ting; Guo, Feng-Biao

    Investigation of essential genes is significant to comprehend the minimal gene sets of cell and discover potential drug targets. In this study, a novel approach based on multiple homology mapping and machine learning method was introduced to predict essential genes. We focused on 25 bacteria which have characterized essential genes. The predictions yielded the highest area under receiver operating characteristic (ROC) curve (AUC) of 0.9716 through tenfold cross-validation test. Proper features were utilized to construct models to make predictions in distantly related bacteria. The accuracy of predictions was evaluated via the consistency of predictions and known essential genes of target species. The highest AUC of 0.9552 and average AUC of 0.8314 were achieved when making predictions across organisms. An independent dataset from Synechococcus elongatus , which was released recently, was obtained for further assessment of the performance of our model. The AUC score of predictions is 0.7855, which is higher than other methods. This research presents that features obtained by homology mapping uniquely can achieve quite great or even better results than those integrated features. Meanwhile, the work indicates that machine learning-based method can assign more efficient weight coefficients than using empirical formula based on biological knowledge.

  17. Statistical tests for equal predictive ability across multiple forecasting methods

    DEFF Research Database (Denmark)

    Borup, Daniel; Thyrsgaard, Martin

    We develop a multivariate generalization of the Giacomini-White tests for equal conditional predictive ability. The tests are applicable to a mixture of nested and non-nested models, incorporate estimation uncertainty explicitly, and allow for misspecification of the forecasting model as well as ...

  18. Quantitative sensory testing predicts pregabalin efficacy in painful chronic pancreatitis

    NARCIS (Netherlands)

    Olesen, S.S.; Graversen, C.; Bouwense, S.A.W.; Goor, H. van; Wilder-Smith, O.H.G.; Drewes, A.M.

    2013-01-01

    BACKGROUND: A major problem in pain medicine is the lack of knowledge about which treatment suits a specific patient. We tested the ability of quantitative sensory testing to predict the analgesic effect of pregabalin and placebo in patients with chronic pancreatitis. METHODS: Sixty-four patients

  19. Predictive Accuracy of Exercise Stress Testing the Healthy Adult.

    Science.gov (United States)

    Lamont, Linda S.

    1981-01-01

    Exercise stress testing provides information on the aerobic capacity, heart rate, and blood pressure responses to graded exercises of a healthy adult. The reliability of exercise tests as a diagnostic procedure is discussed in relation to sensitivity and specificity and predictive accuracy. (JN)

  20. Glutamatergic model psychoses: prediction error, learning, and inference.

    Science.gov (United States)

    Corlett, Philip R; Honey, Garry D; Krystal, John H; Fletcher, Paul C

    2011-01-01

    Modulating glutamatergic neurotransmission induces alterations in conscious experience that mimic the symptoms of early psychotic illness. We review studies that use intravenous administration of ketamine, focusing on interindividual variability in the profundity of the ketamine experience. We will consider this individual variability within a hypothetical model of brain and cognitive function centered upon learning and inference. Within this model, the brains, neural systems, and even single neurons specify expectations about their inputs and responding to violations of those expectations with new learning that renders future inputs more predictable. We argue that ketamine temporarily deranges this ability by perturbing both the ways in which prior expectations are specified and the ways in which expectancy violations are signaled. We suggest that the former effect is predominantly mediated by NMDA blockade and the latter by augmented and inappropriate feedforward glutamatergic signaling. We suggest that the observed interindividual variability emerges from individual differences in neural circuits that normally underpin the learning and inference processes described. The exact source for that variability is uncertain, although it is likely to arise not only from genetic variation but also from subjects' previous experiences and prior learning. Furthermore, we argue that chronic, unlike acute, NMDA blockade alters the specification of expectancies more profoundly and permanently. Scrutinizing individual differences in the effects of acute and chronic ketamine administration in the context of the Bayesian brain model may generate new insights about the symptoms of psychosis; their underlying cognitive processes and neurocircuitry.

  1. Support vector machine incremental learning triggered by wrongly predicted samples

    Science.gov (United States)

    Tang, Ting-long; Guan, Qiu; Wu, Yi-rong

    2018-05-01

    According to the classic Karush-Kuhn-Tucker (KKT) theorem, at every step of incremental support vector machine (SVM) learning, the newly adding sample which violates the KKT conditions will be a new support vector (SV) and migrate the old samples between SV set and non-support vector (NSV) set, and at the same time the learning model should be updated based on the SVs. However, it is not exactly clear at this moment that which of the old samples would change between SVs and NSVs. Additionally, the learning model will be unnecessarily updated, which will not greatly increase its accuracy but decrease the training speed. Therefore, how to choose the new SVs from old sets during the incremental stages and when to process incremental steps will greatly influence the accuracy and efficiency of incremental SVM learning. In this work, a new algorithm is proposed to select candidate SVs and use the wrongly predicted sample to trigger the incremental processing simultaneously. Experimental results show that the proposed algorithm can achieve good performance with high efficiency, high speed and good accuracy.

  2. Global discriminative learning for higher-accuracy computational gene prediction.

    Directory of Open Access Journals (Sweden)

    Axel Bernal

    2007-03-01

    Full Text Available Most ab initio gene predictors use a probabilistic sequence model, typically a hidden Markov model, to combine separately trained models of genomic signals and content. By combining separate models of relevant genomic features, such gene predictors can exploit small training sets and incomplete annotations, and can be trained fairly efficiently. However, that type of piecewise training does not optimize prediction accuracy and has difficulty in accounting for statistical dependencies among different parts of the gene model. With genomic information being created at an ever-increasing rate, it is worth investigating alternative approaches in which many different types of genomic evidence, with complex statistical dependencies, can be integrated by discriminative learning to maximize annotation accuracy. Among discriminative learning methods, large-margin classifiers have become prominent because of the success of support vector machines (SVM in many classification tasks. We describe CRAIG, a new program for ab initio gene prediction based on a conditional random field model with semi-Markov structure that is trained with an online large-margin algorithm related to multiclass SVMs. Our experiments on benchmark vertebrate datasets and on regions from the ENCODE project show significant improvements in prediction accuracy over published gene predictors that use intrinsic features only, particularly at the gene level and on genes with long introns.

  3. Predicting the time of conversion to MCI in the elderly: role of verbal expression and learning.

    Science.gov (United States)

    Oulhaj, Abderrahim; Wilcock, Gordon K; Smith, A David; de Jager, Celeste A

    2009-11-03

    Increasing awareness that minimal or mild cognitive impairment (MCI) in the elderly may be a precursor of dementia has led to an increase in the number of people attending memory clinics. We aimed to develop a way of predicting the period of time before cognitive impairment occurs in community-dwelling elderly. The method is illustrated by the use of simple tests of different cognitive domains. A cohort of 241 normal elderly volunteers was followed for up to 20 years with regular assessments of cognitive abilities using the Cambridge Cognitive Examination (CAMCOG); 91 participants developed MCI. We used interval-censored survival analysis statistical methods to model which baseline cognitive tests best predicted the time to convert to MCI. Out of several baseline variables, only age and CAMCOG subscores for expression and learning/memory were predictors of the time to conversion. The time to conversion was 14% shorter for each 5 years of age, 17% shorter for each point lower in the expression score, and 15% shorter for each point lower in the learning score. We present in tabular form the probability of converting to MCI over intervals between 2 and 10 years for different combinations of expression and learning scores. In apparently normal elderly people, subtle measurable cognitive deficits that occur within the normal range on standard testing protocols reliably predict the time to clinically relevant cognitive impairment long before clinical symptoms are reported.

  4. Lessons Learned in Software Testing A Context-Driven Approach

    CERN Document Server

    Kaner, Cem; Pettichord, Bret

    2008-01-01

    Decades of software testing experience condensed into the most important lessons learned.The world's leading software testing experts lend you their wisdom and years of experience to help you avoid the most common mistakes in testing software. Each lesson is an assertion related to software testing, followed by an explanation or example that shows you the how, when, and why of the testing lesson. More than just tips, tricks, and pitfalls to avoid, Lessons Learned in Software Testing speeds you through the critical testing phase of the software development project without the extensive trial an

  5. Lessons learned in testing of Safeguards equipment

    International Nuclear Information System (INIS)

    Pepper, Susan; Farnitano, Michael; Carelli, Joseph

    2001-01-01

    Upgrade Travel Funding' - This subtask provides funding for the upgrade of DIS equipment installed in the field; E.125.3, 'DIS Radiation Field Characterization' - This subtask provides for the procurement by the IAEA of radiation measurement equipment and technical assistance for the characterization of radiation conditions in the locations where DIS will be installed. This will help the IAEA ensure that the design specifications for the equipment are consistent with the location where the instrument will be used; E.125.4, 'DIS Design Limit Testing and Advise to Strengthen IAEA's Current Equipment Qualification Criteria' - Under this subtask, Wyle Laboratories and Quanterion Solutions will conduct SDIS design limit testing, including harsh environmental testing and accelerated aging, to determine the expected lifetime and produce a design limit report to include maximum operating environment vs. design limit analysis. Additionally, this task will include the development of a strengthened environmental qualification test plan and reliability and maintainability definition methodology for all safeguards equipment. The implementation of new equipment by the Department of Safeguards is costly. Expected costs associated with the implementation of equipment include capital costs, training and in some cases travel. The cost is dramatically increased when operational issues arise due to the costs of studying the issues, modifying and upgrading the equipment and additional travel. The U.S. Support Program believes that the IAEA's Division of Safeguards Technical Support (SGTS) must strengthen its equipment-testing program to ensure that the equipment it approves for inspection use is reliable and will not place additional burden on the Department of Safeguards' maintenance and inspection staff. The U.S. Support Program recognizes that SGTS already requires a series of fundamentally important and revealing tests, but we believe that additional tests should be added to the testing

  6. A Symbiotic Framework for coupling Machine Learning and Geosciences in Prediction and Predictability

    Science.gov (United States)

    Ravela, S.

    2017-12-01

    In this presentation we review the two directions of a symbiotic relationship between machine learning and the geosciences in relation to prediction and predictability. In the first direction, we develop ensemble, information theoretic and manifold learning framework to adaptively improve state and parameter estimates in nonlinear high-dimensional non-Gaussian problems, showing in particular that tractable variational approaches can be produced. We demonstrate these applications in the context of autonomous mapping of environmental coherent structures and other idealized problems. In the reverse direction, we show that data assimilation, particularly probabilistic approaches for filtering and smoothing offer a novel and useful way to train neural networks, and serve as a better basis than gradient based approaches when we must quantify uncertainty in association with nonlinear, chaotic processes. In many inference problems in geosciences we seek to build reduced models to characterize local sensitivies, adjoints or other mechanisms that propagate innovations and errors. Here, the particular use of neural approaches for such propagation trained using ensemble data assimilation provides a novel framework. Through these two examples of inference problems in the earth sciences, we show that not only is learning useful to broaden existing methodology, but in reverse, geophysical methodology can be used to influence paradigms in learning.

  7. Predicting Freshman Grade Point Average From College Admissions Test Scores and State High School Test Scores

    OpenAIRE

    Koretz, Daniel; Yu, C; Mbekeani, Preeya Pandya; Langi, M.; Dhaliwal, Tasminda Kaur; Braslow, David Arthur

    2016-01-01

    The current focus on assessing “college and career readiness” raises an empirical question: How do high school tests compare with college admissions tests in predicting performance in college? We explored this using data from the City University of New York and public colleges in Kentucky. These two systems differ in the choice of college admissions test, the stakes for students on the high school test, and demographics. We predicted freshman grade point average (FGPA) from high school GPA an...

  8. Deep-Learning-Based Drug-Target Interaction Prediction.

    Science.gov (United States)

    Wen, Ming; Zhang, Zhimin; Niu, Shaoyu; Sha, Haozhi; Yang, Ruihan; Yun, Yonghuan; Lu, Hongmei

    2017-04-07

    Identifying interactions between known drugs and targets is a major challenge in drug repositioning. In silico prediction of drug-target interaction (DTI) can speed up the expensive and time-consuming experimental work by providing the most potent DTIs. In silico prediction of DTI can also provide insights about the potential drug-drug interaction and promote the exploration of drug side effects. Traditionally, the performance of DTI prediction depends heavily on the descriptors used to represent the drugs and the target proteins. In this paper, to accurately predict new DTIs between approved drugs and targets without separating the targets into different classes, we developed a deep-learning-based algorithmic framework named DeepDTIs. It first abstracts representations from raw input descriptors using unsupervised pretraining and then applies known label pairs of interaction to build a classification model. Compared with other methods, it is found that DeepDTIs reaches or outperforms other state-of-the-art methods. The DeepDTIs can be further used to predict whether a new drug targets to some existing targets or whether a new target interacts with some existing drugs.

  9. Research on cardiovascular disease prediction based on distance metric learning

    Science.gov (United States)

    Ni, Zhuang; Liu, Kui; Kang, Guixia

    2018-04-01

    Distance metric learning algorithm has been widely applied to medical diagnosis and exhibited its strengths in classification problems. The k-nearest neighbour (KNN) is an efficient method which treats each feature equally. The large margin nearest neighbour classification (LMNN) improves the accuracy of KNN by learning a global distance metric, which did not consider the locality of data distributions. In this paper, we propose a new distance metric algorithm adopting cosine metric and LMNN named COS-SUBLMNN which takes more care about local feature of data to overcome the shortage of LMNN and improve the classification accuracy. The proposed methodology is verified on CVDs patient vector derived from real-world medical data. The Experimental results show that our method provides higher accuracy than KNN and LMNN did, which demonstrates the effectiveness of the Risk predictive model of CVDs based on COS-SUBLMNN.

  10. Distributed Learning, Recognition, and Prediction by ART and ARTMAP Neural Networks.

    Science.gov (United States)

    Carpenter, Gail A.

    1997-11-01

    A class of adaptive resonance theory (ART) models for learning, recognition, and prediction with arbitrarily distributed code representations is introduced. Distributed ART neural networks combine the stable fast learning capabilities of winner-take-all ART systems with the noise tolerance and code compression capabilities of multilayer perceptrons. With a winner-take-all code, the unsupervised model dART reduces to fuzzy ART and the supervised model dARTMAP reduces to fuzzy ARTMAP. With a distributed code, these networks automatically apportion learned changes according to the degree of activation of each coding node, which permits fast as well as slow learning without catastrophic forgetting. Distributed ART models replace the traditional neural network path weight with a dynamic weight equal to the rectified difference between coding node activation and an adaptive threshold. Thresholds increase monotonically during learning according to a principle of atrophy due to disuse. However, monotonic change at the synaptic level manifests itself as bidirectional change at the dynamic level, where the result of adaptation resembles long-term potentiation (LTP) for single-pulse or low frequency test inputs but can resemble long-term depression (LTD) for higher frequency test inputs. This paradoxical behavior is traced to dual computational properties of phasic and tonic coding signal components. A parallel distributed match-reset-search process also helps stabilize memory. Without the match-reset-search system, dART becomes a type of distributed competitive learning network.

  11. Machine Learning Algorithms Outperform Conventional Regression Models in Predicting Development of Hepatocellular Carcinoma

    Science.gov (United States)

    Singal, Amit G.; Mukherjee, Ashin; Elmunzer, B. Joseph; Higgins, Peter DR; Lok, Anna S.; Zhu, Ji; Marrero, Jorge A; Waljee, Akbar K

    2015-01-01

    Background Predictive models for hepatocellular carcinoma (HCC) have been limited by modest accuracy and lack of validation. Machine learning algorithms offer a novel methodology, which may improve HCC risk prognostication among patients with cirrhosis. Our study's aim was to develop and compare predictive models for HCC development among cirrhotic patients, using conventional regression analysis and machine learning algorithms. Methods We enrolled 442 patients with Child A or B cirrhosis at the University of Michigan between January 2004 and September 2006 (UM cohort) and prospectively followed them until HCC development, liver transplantation, death, or study termination. Regression analysis and machine learning algorithms were used to construct predictive models for HCC development, which were tested on an independent validation cohort from the Hepatitis C Antiviral Long-term Treatment against Cirrhosis (HALT-C) Trial. Both models were also compared to the previously published HALT-C model. Discrimination was assessed using receiver operating characteristic curve analysis and diagnostic accuracy was assessed with net reclassification improvement and integrated discrimination improvement statistics. Results After a median follow-up of 3.5 years, 41 patients developed HCC. The UM regression model had a c-statistic of 0.61 (95%CI 0.56-0.67), whereas the machine learning algorithm had a c-statistic of 0.64 (95%CI 0.60–0.69) in the validation cohort. The machine learning algorithm had significantly better diagnostic accuracy as assessed by net reclassification improvement (pmachine learning algorithm (p=0.047). Conclusion Machine learning algorithms improve the accuracy of risk stratifying patients with cirrhosis and can be used to accurately identify patients at high-risk for developing HCC. PMID:24169273

  12. SU-F-P-20: Predicting Waiting Times in Radiation Oncology Using Machine Learning

    International Nuclear Information System (INIS)

    Joseph, A; Herrera, D; Hijal, T; Kildea, J; Hendren, L; Leung, A; Wainberg, J; Sawaf, M; Gorshkov, M; Maglieri, R; Keshavarz, M

    2016-01-01

    Purpose: Waiting times remain one of the most vexing patient satisfaction challenges facing healthcare. Waiting time uncertainty can cause patients, who are already sick or in pain, to worry about when they will receive the care they need. These waiting periods are often difficult for staff to predict and only rough estimates are typically provided based on personal experience. This level of uncertainty leaves most patients unable to plan their calendar, making the waiting experience uncomfortable, even painful. In the present era of electronic health records (EHRs), waiting times need not be so uncertain. Extensive EHRs provide unprecedented amounts of data that can statistically cluster towards representative values when appropriate patient cohorts are selected. Predictive modelling, such as machine learning, is a powerful approach that benefits from large, potentially complex, datasets. The essence of machine learning is to predict future outcomes by learning from previous experience. The application of a machine learning algorithm to waiting time data has the potential to produce personalized waiting time predictions such that the uncertainty may be removed from the patient’s waiting experience. Methods: In radiation oncology, patients typically experience several types of waiting (eg waiting at home for treatment planning, waiting in the waiting room for oncologist appointments and daily waiting in the waiting room for radiotherapy treatments). A daily treatment wait time model is discussed in this report. To develop a prediction model using our large dataset (with more than 100k sample points) a variety of machine learning algorithms from the Python package sklearn were tested. Results: We found that the Random Forest Regressor model provides the best predictions for daily radiotherapy treatment waiting times. Using this model, we achieved a median residual (actual value minus predicted value) of 0.25 minutes and a standard deviation residual of 6.5 minutes

  13. SU-F-P-20: Predicting Waiting Times in Radiation Oncology Using Machine Learning

    Energy Technology Data Exchange (ETDEWEB)

    Joseph, A; Herrera, D; Hijal, T; Kildea, J [McGill University Health Centre, Montreal, Quebec (Canada); Hendren, L; Leung, A; Wainberg, J; Sawaf, M; Gorshkov, M; Maglieri, R; Keshavarz, M [McGill University, Montreal, Quebec (Canada)

    2016-06-15

    Purpose: Waiting times remain one of the most vexing patient satisfaction challenges facing healthcare. Waiting time uncertainty can cause patients, who are already sick or in pain, to worry about when they will receive the care they need. These waiting periods are often difficult for staff to predict and only rough estimates are typically provided based on personal experience. This level of uncertainty leaves most patients unable to plan their calendar, making the waiting experience uncomfortable, even painful. In the present era of electronic health records (EHRs), waiting times need not be so uncertain. Extensive EHRs provide unprecedented amounts of data that can statistically cluster towards representative values when appropriate patient cohorts are selected. Predictive modelling, such as machine learning, is a powerful approach that benefits from large, potentially complex, datasets. The essence of machine learning is to predict future outcomes by learning from previous experience. The application of a machine learning algorithm to waiting time data has the potential to produce personalized waiting time predictions such that the uncertainty may be removed from the patient’s waiting experience. Methods: In radiation oncology, patients typically experience several types of waiting (eg waiting at home for treatment planning, waiting in the waiting room for oncologist appointments and daily waiting in the waiting room for radiotherapy treatments). A daily treatment wait time model is discussed in this report. To develop a prediction model using our large dataset (with more than 100k sample points) a variety of machine learning algorithms from the Python package sklearn were tested. Results: We found that the Random Forest Regressor model provides the best predictions for daily radiotherapy treatment waiting times. Using this model, we achieved a median residual (actual value minus predicted value) of 0.25 minutes and a standard deviation residual of 6.5 minutes

  14. Test-Retest Reliability and Predictive Validity of the Implicit Association Test in Children

    Science.gov (United States)

    Rae, James R.; Olson, Kristina R.

    2018-01-01

    The Implicit Association Test (IAT) is increasingly used in developmental research despite minimal evidence of whether children's IAT scores are reliable across time or predictive of behavior. When test-retest reliability and predictive validity have been assessed, the results have been mixed, and because these studies have differed on many…

  15. Predict-share-observe-explain learning activity for the Torricelli's tank experiment

    Science.gov (United States)

    Panich, Charunya; Puttharugsa, Chokchai; Khemmani, Supitch

    2018-01-01

    The purpose of this research was to study the students' scientific concept and achievement on fluid mechanics before and after the predict-share-observe-explain (PSOE) learning activity for the Torricelli's tank experiment. The 24 participants, who were selected by purposive sampling, were students at grade 12 at Nannakorn School, Nan province. A one group pre-test/post-test design was employed in the study. The research instruments were 1) the lesson plans using the PSOE learning activity and 2) two-tier multiple choice question and subjective tests. The results indicated that students had better scientific concept about Torricelli's tank experiment and the post-test mean score was significantly higher than the pre-test mean score at a 0.05 level of significance. Moreover, the students had retention of knowledge after the PSOE learning activity for 4 weeks at a 0.05 level of significance. The study showed that the PSOE learning activity is suitable for developing students' scientific concept and achievement.

  16. Toetsen als Leerinterventie. Samenvatten in het Testing Effect Paradigma [Tests as learning interventions. Summarization in the testing effect paradigma investigated

    NARCIS (Netherlands)

    Dirkx, Kim; Kester, Liesbeth; Kirschner, Paul A.

    2011-01-01

    Dirkx, K. J. H., Kester, L., & Kirschner, P. A. (2011, July). Toetsen als leerinterventie. Samenvatten in het testing effect paradigma onderzocht [Tests as learning interventions. Summarization in the testing effect paradigma investigated]. Presentation for Erasmus University Rotterdam, Rotterdam.

  17. Putting the Testing Effect to the Test. Why and When is Testing effective for Learning in Secondary School

    NARCIS (Netherlands)

    Dirkx, Kim

    2014-01-01

    Dirkx, K. J. H. (2014, 11 April). Putting the testing effect to the test. Why and when is testing effective for learning in secondary school. Unpublished doctoral dissertation. Heerlen: Open University of the Netherlands

  18. Spam comments prediction using stacking with ensemble learning

    Science.gov (United States)

    Mehmood, Arif; On, Byung-Won; Lee, Ingyu; Ashraf, Imran; Choi, Gyu Sang

    2018-01-01

    Illusive comments of product or services are misleading for people in decision making. The current methodologies to predict deceptive comments are concerned for feature designing with single training model. Indigenous features have ability to show some linguistic phenomena but are hard to reveal the latent semantic meaning of the comments. We propose a prediction model on general features of documents using stacking with ensemble learning. Term Frequency/Inverse Document Frequency (TF/IDF) features are inputs to stacking of Random Forest and Gradient Boosted Trees and the outputs of the base learners are encapsulated with decision tree to make final training of the model. The results exhibits that our approach gives the accuracy of 92.19% which outperform the state-of-the-art method.

  19. Negative correlation learning for customer churn prediction: a comparison study.

    Science.gov (United States)

    Rodan, Ali; Fayyoumi, Ayham; Faris, Hossam; Alsakran, Jamal; Al-Kadi, Omar

    2015-01-01

    Recently, telecommunication companies have been paying more attention toward the problem of identification of customer churn behavior. In business, it is well known for service providers that attracting new customers is much more expensive than retaining existing ones. Therefore, adopting accurate models that are able to predict customer churn can effectively help in customer retention campaigns and maximizing the profit. In this paper we will utilize an ensemble of Multilayer perceptrons (MLP) whose training is obtained using negative correlation learning (NCL) for predicting customer churn in a telecommunication company. Experiments results confirm that NCL based MLP ensemble can achieve better generalization performance (high churn rate) compared with ensemble of MLP without NCL (flat ensemble) and other common data mining techniques used for churn analysis.

  20. Predictive testing of performance of metals in HTR service environments

    International Nuclear Information System (INIS)

    Kondo, T.; Shindo, M.; Tamura, M.; Tsuji, H.; Kurata, Y.; Tsukada, T.

    1982-01-01

    Status of the material testing in simulated HTGR environment is reviewed with special attention focused on the methodology of the prediction of performance in long time. Importance of controlling effective chemical potentials relations in the material-environmental interface is stressed in regard of the complex inter-dependent kinetic relation between oxidation and carbon transport. Based on the recent experimental observations, proposals are made to establish some procedures for conservative prediction of the metal performance

  1. Days on radiosensitivity: individual variability and predictive tests

    International Nuclear Information System (INIS)

    2008-01-01

    The radiosensitivity is a part of usual clinical observations. It is already included in the therapy protocols. however, some questions stay on its individual variability and on the difficulty to evaluate it. The point will be stocked on its origin and its usefulness in predictive medicine. Through examples on the use of predictive tests and ethical and legal questions that they raise, concrete cases will be presented by specialists such radio biologists, geneticists, immunologists, jurists and occupational physicians. (N.C.)

  2. Learning Latent Variable and Predictive Models of Dynamical Systems

    Science.gov (United States)

    2009-10-01

    Huijbregts. The ICSI RT07s Speaker Diarization System. Springer-Verlag, 2008. 4.5 [57] Gal Elidan and Nir Friedman. Learning the dimensionality of hidden...13, 435 and a test set of size 1, 771. VOWEL: This data set consists of multiple utterances of a particular Japanese vowel by nine male speakers . We...classification based on cultural style [51]; audio diarization , i.e., extraction of speech segments in long audio signals from background sounds [52]; audio

  3. Role of Personality Traits, Learning Styles and Metacognition in Predicting Critical Thinking of Undergraduate Students

    Directory of Open Access Journals (Sweden)

    Soliemanifar O

    2015-04-01

    The aim of this study was to investigate the role of personality traits, learning styles and metacognition in predicting critical thinking. Instrument & Methods: In this descriptive correlative study, 240 students (130 girls and 110 boys of Ahvaz Shahid Chamran University were selected by multi-stage random sampling method. The instruments for collecting data were NEO Five-Factor Inventory, learning style inventory of Kolb (LSI, metacognitive assessment inventory (MAI of Schraw & Dennison (1994 and California Critical Thinking Skills Test (CCTST. The data were analyzed using Pearson correlation coefficient, stepwise regression analysis and Canonical correlation analysis.  Findings: Openness to experiment (b=0.41, conscientiousness (b=0.28, abstract conceptualization (b=0.39, active experimentation (b=0.22, reflective observation (b=0.12, knowledge of cognition (b=0.47 and regulation of cognition (b=0.29 were effective in predicting critical thinking. Openness to experiment and conscientiousness (r2=0.25, active experimentation, abstract conceptualization and reflective observation learning styles (r2=0.21 and knowledge and regulation of cognition metacognitions (r2=0.3 had an important role in explaining critical thinking. The linear combination of critical thinking skills (evaluation, analysis, inference was predictable by a linear combination of dispositional-cognitive factors (openness, conscientiousness, abstract conceptualization, active experimentation, knowledge of cognition and regulation of cognition. Conclusion: Personality traits, learning styles and metacognition, as dispositional-cognitive factors, play a significant role in students' critical thinking.

  4. Efficient Prediction of Low-Visibility Events at Airports Using Machine-Learning Regression

    Science.gov (United States)

    Cornejo-Bueno, L.; Casanova-Mateo, C.; Sanz-Justo, J.; Cerro-Prada, E.; Salcedo-Sanz, S.

    2017-11-01

    We address the prediction of low-visibility events at airports using machine-learning regression. The proposed model successfully forecasts low-visibility events in terms of the runway visual range at the airport, with the use of support-vector regression, neural networks (multi-layer perceptrons and extreme-learning machines) and Gaussian-process algorithms. We assess the performance of these algorithms based on real data collected at the Valladolid airport, Spain. We also propose a study of the atmospheric variables measured at a nearby tower related to low-visibility atmospheric conditions, since they are considered as the inputs of the different regressors. A pre-processing procedure of these input variables with wavelet transforms is also described. The results show that the proposed machine-learning algorithms are able to predict low-visibility events well. The Gaussian process is the best algorithm among those analyzed, obtaining over 98% of the correct classification rate in low-visibility events when the runway visual range is {>}1000 m, and about 80% under this threshold. The performance of all the machine-learning algorithms tested is clearly affected in extreme low-visibility conditions ({algorithm performance in daytime and nighttime conditions, and for different prediction time horizons.

  5. Prediction of outcome in internet-delivered cognitive behaviour therapy for paediatric obsessive-compulsive disorder: A machine learning approach.

    Science.gov (United States)

    Lenhard, Fabian; Sauer, Sebastian; Andersson, Erik; Månsson, Kristoffer Nt; Mataix-Cols, David; Rück, Christian; Serlachius, Eva

    2018-03-01

    There are no consistent predictors of treatment outcome in paediatric obsessive-compulsive disorder (OCD). One reason for this might be the use of suboptimal statistical methodology. Machine learning is an approach to efficiently analyse complex data. Machine learning has been widely used within other fields, but has rarely been tested in the prediction of paediatric mental health treatment outcomes. To test four different machine learning methods in the prediction of treatment response in a sample of paediatric OCD patients who had received Internet-delivered cognitive behaviour therapy (ICBT). Participants were 61 adolescents (12-17 years) who enrolled in a randomized controlled trial and received ICBT. All clinical baseline variables were used to predict strictly defined treatment response status three months after ICBT. Four machine learning algorithms were implemented. For comparison, we also employed a traditional logistic regression approach. Multivariate logistic regression could not detect any significant predictors. In contrast, all four machine learning algorithms performed well in the prediction of treatment response, with 75 to 83% accuracy. The results suggest that machine learning algorithms can successfully be applied to predict paediatric OCD treatment outcome. Validation studies and studies in other disorders are warranted. Copyright © 2017 John Wiley & Sons, Ltd.

  6. Cognitive Learning Strategy as a Partial Effect on Major Field Test in Business Results

    Science.gov (United States)

    Strang, Kenneth David

    2014-01-01

    An experiment was developed to determine if cognitive learning strategies improved standardized university business exam results. Previous studies revealed that factors such as prior ability, age, gender, and culture predicted a student's Major Field Test in Business (MFTB) score better than course content. The experiment control consisted of…

  7. Help-Seeking Decisions of Battered Women: A Test of Learned Helplessness and Two Stress Theories.

    Science.gov (United States)

    Wauchope, Barbara A.

    This study tested the learned helplessness theory, stress theory, and a modified stress theory to determine the best model for predicting the probability that a woman would seek help when she experienced severe violence from a male partner. The probability was hypothesized to increase as the stress of the violence experienced increased. Data were…

  8. Epileptic Seizure Prediction Using Big Data and Deep Learning: Toward a Mobile System.

    Science.gov (United States)

    Kiral-Kornek, Isabell; Roy, Subhrajit; Nurse, Ewan; Mashford, Benjamin; Karoly, Philippa; Carroll, Thomas; Payne, Daniel; Saha, Susmita; Baldassano, Steven; O'Brien, Terence; Grayden, David; Cook, Mark; Freestone, Dean; Harrer, Stefan

    2018-01-01

    Seizure prediction can increase independence and allow preventative treatment for patients with epilepsy. We present a proof-of-concept for a seizure prediction system that is accurate, fully automated, patient-specific, and tunable to an individual's needs. Intracranial electroencephalography (iEEG) data of ten patients obtained from a seizure advisory system were analyzed as part of a pseudoprospective seizure prediction study. First, a deep learning classifier was trained to distinguish between preictal and interictal signals. Second, classifier performance was tested on held-out iEEG data from all patients and benchmarked against the performance of a random predictor. Third, the prediction system was tuned so sensitivity or time in warning could be prioritized by the patient. Finally, a demonstration of the feasibility of deployment of the prediction system onto an ultra-low power neuromorphic chip for autonomous operation on a wearable device is provided. The prediction system achieved mean sensitivity of 69% and mean time in warning of 27%, significantly surpassing an equivalent random predictor for all patients by 42%. This study demonstrates that deep learning in combination with neuromorphic hardware can provide the basis for a wearable, real-time, always-on, patient-specific seizure warning system with low power consumption and reliable long-term performance. Copyright © 2017 The Authors. Published by Elsevier B.V. All rights reserved.

  9. Positive-Unlabeled Learning for Pupylation Sites Prediction

    Directory of Open Access Journals (Sweden)

    Ming Jiang

    2016-01-01

    Full Text Available Pupylation plays a key role in regulating various protein functions as a crucial posttranslational modification of prokaryotes. In order to understand the molecular mechanism of pupylation, it is important to identify pupylation substrates and sites accurately. Several computational methods have been developed to identify pupylation sites because the traditional experimental methods are time-consuming and labor-sensitive. With the existing computational methods, the experimentally annotated pupylation sites are used as the positive training set and the remaining nonannotated lysine residues as the negative training set to build classifiers to predict new pupylation sites from the unknown proteins. However, the remaining nonannotated lysine residues may contain pupylation sites which have not been experimentally validated yet. Unlike previous methods, in this study, the experimentally annotated pupylation sites were used as the positive training set whereas the remaining nonannotated lysine residues were used as the unlabeled training set. A novel method named PUL-PUP was proposed to predict pupylation sites by using positive-unlabeled learning technique. Our experimental results indicated that PUL-PUP outperforms the other methods significantly for the prediction of pupylation sites. As an application, PUL-PUP was also used to predict the most likely pupylation sites in nonannotated lysine sites.

  10. Team-Based Testing Improves Individual Learning

    Science.gov (United States)

    Vogler, Jane S.; Robinson, Daniel H.

    2016-01-01

    In two experiments, 90 undergraduates took six tests as part of an educational psychology course. Using a crossover design, students took three tests individually without feedback and then took the same test again, following the process of team-based testing (TBT), in teams in which the members reached consensus for each question and answered…

  11. A data-driven predictive approach for drug delivery using machine learning techniques.

    Directory of Open Access Journals (Sweden)

    Yuanyuan Li

    Full Text Available In drug delivery, there is often a trade-off between effective killing of the pathogen, and harmful side effects associated with the treatment. Due to the difficulty in testing every dosing scenario experimentally, a computational approach will be helpful to assist with the prediction of effective drug delivery methods. In this paper, we have developed a data-driven predictive system, using machine learning techniques, to determine, in silico, the effectiveness of drug dosing. The system framework is scalable, autonomous, robust, and has the ability to predict the effectiveness of the current drug treatment and the subsequent drug-pathogen dynamics. The system consists of a dynamic model incorporating both the drug concentration and pathogen population into distinct states. These states are then analyzed using a temporal model to describe the drug-cell interactions over time. The dynamic drug-cell interactions are learned in an adaptive fashion and used to make sequential predictions on the effectiveness of the dosing strategy. Incorporated into the system is the ability to adjust the sensitivity and specificity of the learned models based on a threshold level determined by the operator for the specific application. As a proof-of-concept, the system was validated experimentally using the pathogen Giardia lamblia and the drug metronidazole in vitro.

  12. Application of Machine Learning for Dragline Failure Prediction

    Directory of Open Access Journals (Sweden)

    Taghizadeh Amir

    2017-01-01

    Full Text Available Overburden stripping in open cast coal mines is extensively carried out by walking draglines. Draglines’ unavailability and unexpected failures result in delayed productions and increased maintenance and operating costs. Therefore, achieving high availability of draglines plays a crucial role for increasing economic feasibility of mining projects. Applications of methodologies which can forecast the failure type of dragline based on the available failure data not only help to reduce the maintenance and operating costs but also increase the availability and the production rate. In this study, Machine Learning approaches have been applied for data which has been gathered from an operating coal mine in Turkey. The study methodology consists of three algorithms as: i implementation of K-Nearest Neighbors, ii implementation of Multi-Layer Perceptron, and iii implementation of Radial Basis Function. The algorithms have been utilized for predicting the draglines’ failure types. In this sense, the input data, which are mean time-to-failure, and the output data, failure types, have been fed to the algorithms. The regression analysis of methodologies have been compared and showed the K- Nearest Neighbors has a higher rate of regression which is around 70 percent. Thus, the K-Nearest Neighbor algorithm can be applied in order to preventive components replacement which causes to minimized preventive and corrective cost parameters. The accurate prediction of failure type, indeed, causes to optimized number of inspections. The novelty of this study is application of machine learning approaches in draglines’ reliability subject for first time.

  13. Application of Machine Learning Algorithms for the Query Performance Prediction

    Directory of Open Access Journals (Sweden)

    MILICEVIC, M.

    2015-08-01

    Full Text Available This paper analyzes the relationship between the system load/throughput and the query response time in a real Online transaction processing (OLTP system environment. Although OLTP systems are characterized by short transactions, which normally entail high availability and consistent short response times, the need for operational reporting may jeopardize these objectives. We suggest a new approach to performance prediction for concurrent database workloads, based on the system state vector which consists of 36 attributes. There is no bias to the importance of certain attributes, but the machine learning methods are used to determine which attributes better describe the behavior of the particular database server and how to model that system. During the learning phase, the system's profile is created using multiple reference queries, which are selected to represent frequent business processes. The possibility of the accurate response time prediction may be a foundation for automated decision-making for database (DB query scheduling. Possible applications of the proposed method include adaptive resource allocation, quality of service (QoS management or real-time dynamic query scheduling (e.g. estimation of the optimal moment for a complex query execution.

  14. Hierarchical prediction errors in midbrain and septum during social learning.

    Science.gov (United States)

    Diaconescu, Andreea O; Mathys, Christoph; Weber, Lilian A E; Kasper, Lars; Mauer, Jan; Stephan, Klaas E

    2017-04-01

    Social learning is fundamental to human interactions, yet its computational and physiological mechanisms are not well understood. One prominent open question concerns the role of neuromodulatory transmitters. We combined fMRI, computational modelling and genetics to address this question in two separate samples (N = 35, N = 47). Participants played a game requiring inference on an adviser's intentions whose motivation to help or mislead changed over time. Our analyses suggest that hierarchically structured belief updates about current advice validity and the adviser's trustworthiness, respectively, depend on different neuromodulatory systems. Low-level prediction errors (PEs) about advice accuracy not only activated regions known to support 'theory of mind', but also the dopaminergic midbrain. Furthermore, PE responses in ventral striatum were influenced by the Met/Val polymorphism of the Catechol-O-Methyltransferase (COMT) gene. By contrast, high-level PEs ('expected uncertainty') about the adviser's fidelity activated the cholinergic septum. These findings, replicated in both samples, have important implications: They suggest that social learning rests on hierarchically related PEs encoded by midbrain and septum activity, respectively, in the same manner as other forms of learning under volatility. Furthermore, these hierarchical PEs may be broadcast by dopaminergic and cholinergic projections to induce plasticity specifically in cortical areas known to represent beliefs about others. © The Author (2017). Published by Oxford University Press.

  15. Prediction of EMP cavitation threshold from other than sodium testing

    International Nuclear Information System (INIS)

    Kambe, M.; Kamei, M.

    2002-01-01

    An experimental study has been performed to predict the cavitation threshold of electromagnetic pumps from measurements on test models using water and alcohol. Cavitation tests were carried out using water and alcohol test loop on subscale ducts of transparent acrylic resin with reference to an actual pump (1.1m 3 /min). These data were compared to those obtained from the in-sodium tests on the actual pump. The investigation revealed that the value of Thoma's dimensionless parameter: σ applied to the test model for water and alcohol is quite higher than that of corresponding σ on the actual pump. To minimize the incipient cavitation safety margin, more accurate prediction must be required. In view of this, the authors proposed the dimensionless parameter: σ T =σ/W-bare where W-bare denotes the Weber number. This parameter was confirmed to predict the cavitation threshold of electromagnetic pumps with much more accuracy than ever before. It can also be adopted to predict cavitation threshold of other FBR components. (author)

  16. Pediatric residents' learning styles and temperaments and their relationships to standardized test scores.

    Science.gov (United States)

    Tuli, Sanjeev Y; Thompson, Lindsay A; Saliba, Heidi; Black, Erik W; Ryan, Kathleen A; Kelly, Maria N; Novak, Maureen; Mellott, Jane; Tuli, Sonal S

    2011-12-01

    Board certification is an important professional qualification and a prerequisite for credentialing, and the Accreditation Council for Graduate Medical Education (ACGME) assesses board certification rates as a component of residency program effectiveness. To date, research has shown that preresidency measures, including National Board of Medical Examiners scores, Alpha Omega Alpha Honor Medical Society membership, or medical school grades poorly predict postresidency board examination scores. However, learning styles and temperament have been identified as factors that 5 affect test-taking performance. The purpose of this study is to characterize the learning styles and temperaments of pediatric residents and to evaluate their relationships to yearly in-service and postresidency board examination scores. This cross-sectional study analyzed the learning styles and temperaments of current and past pediatric residents by administration of 3 validated tools: the Kolb Learning Style Inventory, the Keirsey Temperament Sorter, and the Felder-Silverman Learning Style test. These results were compared with known, normative, general and medical population data and evaluated for correlation to in-service examination and postresidency board examination scores. The predominant learning style for pediatric residents was converging 44% (33 of 75 residents) and the predominant temperament was guardian 61% (34 of 56 residents). The learning style and temperament distribution of the residents was significantly different from published population data (P  =  .002 and .04, respectively). Learning styles, with one exception, were found to be unrelated to standardized test scores. The predominant learning style and temperament of pediatric residents is significantly different than that of the populations of general and medical trainees. However, learning styles and temperament do not predict outcomes on standardized in-service and board examinations in pediatric residents.

  17. Curiosity and reward: Valence predicts choice and information prediction errors enhance learning.

    Science.gov (United States)

    Marvin, Caroline B; Shohamy, Daphna

    2016-03-01

    Curiosity drives many of our daily pursuits and interactions; yet, we know surprisingly little about how it works. Here, we harness an idea implied in many conceptualizations of curiosity: that information has value in and of itself. Reframing curiosity as the motivation to obtain reward-where the reward is information-allows one to leverage major advances in theoretical and computational mechanisms of reward-motivated learning. We provide new evidence supporting 2 predictions that emerge from this framework. First, we find an asymmetric effect of positive versus negative information, with positive information enhancing both curiosity and long-term memory for information. Second, we find that it is not the absolute value of information that drives learning but, rather, the gap between the reward expected and reward received, an "information prediction error." These results support the idea that information functions as a reward, much like money or food, guiding choices and driving learning in systematic ways. (c) 2016 APA, all rights reserved).

  18. Learning Predictive Interactions Using Information Gain and Bayesian Network Scoring.

    Directory of Open Access Journals (Sweden)

    Xia Jiang

    Full Text Available The problems of correlation and classification are long-standing in the fields of statistics and machine learning, and techniques have been developed to address these problems. We are now in the era of high-dimensional data, which is data that can concern billions of variables. These data present new challenges. In particular, it is difficult to discover predictive variables, when each variable has little marginal effect. An example concerns Genome-wide Association Studies (GWAS datasets, which involve millions of single nucleotide polymorphism (SNPs, where some of the SNPs interact epistatically to affect disease status. Towards determining these interacting SNPs, researchers developed techniques that addressed this specific problem. However, the problem is more general, and so these techniques are applicable to other problems concerning interactions. A difficulty with many of these techniques is that they do not distinguish whether a learned interaction is actually an interaction or whether it involves several variables with strong marginal effects.We address this problem using information gain and Bayesian network scoring. First, we identify candidate interactions by determining whether together variables provide more information than they do separately. Then we use Bayesian network scoring to see if a candidate interaction really is a likely model. Our strategy is called MBS-IGain. Using 100 simulated datasets and a real GWAS Alzheimer's dataset, we investigated the performance of MBS-IGain.When analyzing the simulated datasets, MBS-IGain substantially out-performed nine previous methods at locating interacting predictors, and at identifying interactions exactly. When analyzing the real Alzheimer's dataset, we obtained new results and results that substantiated previous findings. We conclude that MBS-IGain is highly effective at finding interactions in high-dimensional datasets. This result is significant because we have increasingly

  19. Computerized adaptive testing item selection in computerized adaptive learning systems

    NARCIS (Netherlands)

    Eggen, Theodorus Johannes Hendrikus Maria; Eggen, T.J.H.M.; Veldkamp, B.P.

    2012-01-01

    Item selection methods traditionally developed for computerized adaptive testing (CAT) are explored for their usefulness in item-based computerized adaptive learning (CAL) systems. While in CAT Fisher information-based selection is optimal, for recovering learning populations in CAL systems item

  20. Testing a Conception of How School Leadership Influences Student Learning

    Science.gov (United States)

    Leithwood, Kenneth; Patten, Sarah; Jantzi, Doris

    2010-01-01

    Purpose: This article describes and reports the results of testing a new conception of how leadership influences student learning ("The Four Paths"). Framework: Leadership influence is conceptualized as flowing along four paths (Rational, Emotions, Organizational, and Family) toward student learning. Each path is populated by multiple…

  1. Deep learning architectures for multi-label classification of intelligent health risk prediction.

    Science.gov (United States)

    Maxwell, Andrew; Li, Runzhi; Yang, Bei; Weng, Heng; Ou, Aihua; Hong, Huixiao; Zhou, Zhaoxian; Gong, Ping; Zhang, Chaoyang

    2017-12-28

    Multi-label classification of data remains to be a challenging problem. Because of the complexity of the data, it is sometimes difficult to infer information about classes that are not mutually exclusive. For medical data, patients could have symptoms of multiple different diseases at the same time and it is important to develop tools that help to identify problems early. Intelligent health risk prediction models built with deep learning architectures offer a powerful tool for physicians to identify patterns in patient data that indicate risks associated with certain types of chronic diseases. Physical examination records of 110,300 anonymous patients were used to predict diabetes, hypertension, fatty liver, a combination of these three chronic diseases, and the absence of disease (8 classes in total). The dataset was split into training (90%) and testing (10%) sub-datasets. Ten-fold cross validation was used to evaluate prediction accuracy with metrics such as precision, recall, and F-score. Deep Learning (DL) architectures were compared with standard and state-of-the-art multi-label classification methods. Preliminary results suggest that Deep Neural Networks (DNN), a DL architecture, when applied to multi-label classification of chronic diseases, produced accuracy that was comparable to that of common methods such as Support Vector Machines. We have implemented DNNs to handle both problem transformation and algorithm adaption type multi-label methods and compare both to see which is preferable. Deep Learning architectures have the potential of inferring more information about the patterns of physical examination data than common classification methods. The advanced techniques of Deep Learning can be used to identify the significance of different features from physical examination data as well as to learn the contributions of each feature that impact a patient's risk for chronic diseases. However, accurate prediction of chronic disease risks remains a challenging

  2. Prediction of software operational reliability using testing environment factors

    International Nuclear Information System (INIS)

    Jung, Hoan Sung; Seong, Poong Hyun

    1995-01-01

    For many years, many researches have focused on the quantification of software reliability and there are many models developed to quantify software reliability. Most software reliability models estimate the reliability with the failure data collected during the test assuming that the test environments well represent the operation profile. The experiences show that the operational reliability is higher than the test reliability User's interest is on the operational reliability rather than on the test reliability, however. With the assumption that the difference in reliability results from the change of environment, testing environment factors comprising the aging factor and the coverage factor are defined in this study to predict the ultimate operational reliability with the failure data. It is by incorporating test environments applied beyond the operational profile into testing environment factors. The application results are close to the actual data

  3. Using machine learning to predict wind turbine power output

    International Nuclear Information System (INIS)

    Clifton, A; Kilcher, L; Lundquist, J K; Fleming, P

    2013-01-01

    Wind turbine power output is known to be a strong function of wind speed, but is also affected by turbulence and shear. In this work, new aerostructural simulations of a generic 1.5 MW turbine are used to rank atmospheric influences on power output. Most significant is the hub height wind speed, followed by hub height turbulence intensity and then wind speed shear across the rotor disk. These simulation data are used to train regression trees that predict the turbine response for any combination of wind speed, turbulence intensity, and wind shear that might be expected at a turbine site. For a randomly selected atmospheric condition, the accuracy of the regression tree power predictions is three times higher than that from the traditional power curve methodology. The regression tree method can also be applied to turbine test data and used to predict turbine performance at a new site. No new data are required in comparison to the data that are usually collected for a wind resource assessment. Implementing the method requires turbine manufacturers to create a turbine regression tree model from test site data. Such an approach could significantly reduce bias in power predictions that arise because of the different turbulence and shear at the new site, compared to the test site. (letter)

  4. Early hospital mortality prediction of intensive care unit patients using an ensemble learning approach.

    Science.gov (United States)

    Awad, Aya; Bader-El-Den, Mohamed; McNicholas, James; Briggs, Jim

    2017-12-01

    Mortality prediction of hospitalized patients is an important problem. Over the past few decades, several severity scoring systems and machine learning mortality prediction models have been developed for predicting hospital mortality. By contrast, early mortality prediction for intensive care unit patients remains an open challenge. Most research has focused on severity of illness scoring systems or data mining (DM) models designed for risk estimation at least 24 or 48h after ICU admission. This study highlights the main data challenges in early mortality prediction in ICU patients and introduces a new machine learning based framework for Early Mortality Prediction for Intensive Care Unit patients (EMPICU). The proposed method is evaluated on the Multiparameter Intelligent Monitoring in Intensive Care II (MIMIC-II) database. Mortality prediction models are developed for patients at the age of 16 or above in Medical ICU (MICU), Surgical ICU (SICU) or Cardiac Surgery Recovery Unit (CSRU). We employ the ensemble learning Random Forest (RF), the predictive Decision Trees (DT), the probabilistic Naive Bayes (NB) and the rule-based Projective Adaptive Resonance Theory (PART) models. The primary outcome was hospital mortality. The explanatory variables included demographic, physiological, vital signs and laboratory test variables. Performance measures were calculated using cross-validated area under the receiver operating characteristic curve (AUROC) to minimize bias. 11,722 patients with single ICU stays are considered. Only patients at the age of 16 years old and above in Medical ICU (MICU), Surgical ICU (SICU) or Cardiac Surgery Recovery Unit (CSRU) are considered in this study. The proposed EMPICU framework outperformed standard scoring systems (SOFA, SAPS-I, APACHE-II, NEWS and qSOFA) in terms of AUROC and time (i.e. at 6h compared to 48h or more after admission). The results show that although there are many values missing in the first few hour of ICU admission

  5. Prediction of preterm deliveries from EHG signals using machine learning.

    Directory of Open Access Journals (Sweden)

    Paul Fergus

    Full Text Available There has been some improvement in the treatment of preterm infants, which has helped to increase their chance of survival. However, the rate of premature births is still globally increasing. As a result, this group of infants are most at risk of developing severe medical conditions that can affect the respiratory, gastrointestinal, immune, central nervous, auditory and visual systems. In extreme cases, this can also lead to long-term conditions, such as cerebral palsy, mental retardation, learning difficulties, including poor health and growth. In the US alone, the societal and economic cost of preterm births, in 2005, was estimated to be $26.2 billion, per annum. In the UK, this value was close to £2.95 billion, in 2009. Many believe that a better understanding of why preterm births occur, and a strategic focus on prevention, will help to improve the health of children and reduce healthcare costs. At present, most methods of preterm birth prediction are subjective. However, a strong body of evidence suggests the analysis of uterine electrical signals (Electrohysterography, could provide a viable way of diagnosing true labour and predict preterm deliveries. Most Electrohysterography studies focus on true labour detection during the final seven days, before labour. The challenge is to utilise Electrohysterography techniques to predict preterm delivery earlier in the pregnancy. This paper explores this idea further and presents a supervised machine learning approach that classifies term and preterm records, using an open source dataset containing 300 records (38 preterm and 262 term. The synthetic minority oversampling technique is used to oversample the minority preterm class, and cross validation techniques, are used to evaluate the dataset against other similar studies. Our approach shows an improvement on existing studies with 96% sensitivity, 90% specificity, and a 95% area under the curve value with 8% global error using the polynomial

  6. Learned Helplessness: The Effect of Failure on Test-Taking

    Science.gov (United States)

    Firmin, Michael; Hwang, Chi-En; Copella, Margaret; Clark, Sarah

    2004-01-01

    This study examined learned helplessness and its effect on test taking. Students were given one of two tests; the first began with extremely difficult questions and the other started with easy questions. The researchers hypothesized that those who took the test beginning with difficult questions would become easily frustrated and possibly doubt…

  7. Testing as an inevitable instrument in today's language learning ...

    African Journals Online (AJOL)

    In the foregoing analysis, we have unequivocably declared that testing or assessment is an inevitable instrument in present day's language learning environment and that testing is the other side of teaching itself. In teaching any aspect of language, learner must be tested to determine the progress made so far as well as ...

  8. Testing the prediction error difference between two predictors

    NARCIS (Netherlands)

    van de Wiel, M.A.; Berkhof, J.; van Wieringen, W.N.

    2009-01-01

    We develop an inference framework for the difference in errors between 2 prediction procedures. The 2 procedures may differ in any aspect and possibly utilize different sets of covariates. We apply training and testing on the same data set, which is accommodated by sample splitting. For each split,

  9. Learning Pulse: Using Wearable Biosensors and Learning Analytics to Investigate and Predict Learning Success in Self-regulated Learning

    NARCIS (Netherlands)

    Di Mitri, Daniele; Scheffel, Maren; Drachsler, Hendrik; Börner, Dirk; Ternier, Stefaan; Specht, Marcus

    2017-01-01

    The Learning Pulse study aims to explore whether physiological data such as heart rate and step count correlate with learning activity data and whether they are good predictors for learning success during self-regulated learning. To verify this hypothesis an experiment was set up involving eight

  10. Supervised machine learning techniques to predict binding affinity. A study for cyclin-dependent kinase 2.

    Science.gov (United States)

    de Ávila, Maurício Boff; Xavier, Mariana Morrone; Pintro, Val Oliveira; de Azevedo, Walter Filgueira

    2017-12-09

    Here we report the development of a machine-learning model to predict binding affinity based on the crystallographic structures of protein-ligand complexes. We used an ensemble of crystallographic structures (resolution better than 1.5 Å resolution) for which half-maximal inhibitory concentration (IC 50 ) data is available. Polynomial scoring functions were built using as explanatory variables the energy terms present in the MolDock and PLANTS scoring functions. Prediction performance was tested and the supervised machine learning models showed improvement in the prediction power, when compared with PLANTS and MolDock scoring functions. In addition, the machine-learning model was applied to predict binding affinity of CDK2, which showed a better performance when compared with AutoDock4, AutoDock Vina, MolDock, and PLANTS scores. Copyright © 2017 Elsevier Inc. All rights reserved.

  11. Are students' impressions of improved learning through active learning methods reflected by improved test scores?

    Science.gov (United States)

    Everly, Marcee C

    2013-02-01

    To report the transformation from lecture to more active learning methods in a maternity nursing course and to evaluate whether student perception of improved learning through active-learning methods is supported by improved test scores. The process of transforming a course into an active-learning model of teaching is described. A voluntary mid-semester survey for student acceptance of the new teaching method was conducted. Course examination results, from both a standardized exam and a cumulative final exam, among students who received lecture in the classroom and students who had active learning activities in the classroom were compared. Active learning activities were very acceptable to students. The majority of students reported learning more from having active-learning activities in the classroom rather than lecture-only and this belief was supported by improved test scores. Students who had active learning activities in the classroom scored significantly higher on a standardized assessment test than students who received lecture only. The findings support the use of student reflection to evaluate the effectiveness of active-learning methods and help validate the use of student reflection of improved learning in other research projects. Copyright © 2011 Elsevier Ltd. All rights reserved.

  12. Open source machine-learning algorithms for the prediction of optimal cancer drug therapies.

    Science.gov (United States)

    Huang, Cai; Mezencev, Roman; McDonald, John F; Vannberg, Fredrik

    2017-01-01

    Precision medicine is a rapidly growing area of modern medical science and open source machine-learning codes promise to be a critical component for the successful development of standardized and automated analysis of patient data. One important goal of precision cancer medicine is the accurate prediction of optimal drug therapies from the genomic profiles of individual patient tumors. We introduce here an open source software platform that employs a highly versatile support vector machine (SVM) algorithm combined with a standard recursive feature elimination (RFE) approach to predict personalized drug responses from gene expression profiles. Drug specific models were built using gene expression and drug response data from the National Cancer Institute panel of 60 human cancer cell lines (NCI-60). The models are highly accurate in predicting the drug responsiveness of a variety of cancer cell lines including those comprising the recent NCI-DREAM Challenge. We demonstrate that predictive accuracy is optimized when the learning dataset utilizes all probe-set expression values from a diversity of cancer cell types without pre-filtering for genes generally considered to be "drivers" of cancer onset/progression. Application of our models to publically available ovarian cancer (OC) patient gene expression datasets generated predictions consistent with observed responses previously reported in the literature. By making our algorithm "open source", we hope to facilitate its testing in a variety of cancer types and contexts leading to community-driven improvements and refinements in subsequent applications.

  13. Open source machine-learning algorithms for the prediction of optimal cancer drug therapies.

    Directory of Open Access Journals (Sweden)

    Cai Huang

    Full Text Available Precision medicine is a rapidly growing area of modern medical science and open source machine-learning codes promise to be a critical component for the successful development of standardized and automated analysis of patient data. One important goal of precision cancer medicine is the accurate prediction of optimal drug therapies from the genomic profiles of individual patient tumors. We introduce here an open source software platform that employs a highly versatile support vector machine (SVM algorithm combined with a standard recursive feature elimination (RFE approach to predict personalized drug responses from gene expression profiles. Drug specific models were built using gene expression and drug response data from the National Cancer Institute panel of 60 human cancer cell lines (NCI-60. The models are highly accurate in predicting the drug responsiveness of a variety of cancer cell lines including those comprising the recent NCI-DREAM Challenge. We demonstrate that predictive accuracy is optimized when the learning dataset utilizes all probe-set expression values from a diversity of cancer cell types without pre-filtering for genes generally considered to be "drivers" of cancer onset/progression. Application of our models to publically available ovarian cancer (OC patient gene expression datasets generated predictions consistent with observed responses previously reported in the literature. By making our algorithm "open source", we hope to facilitate its testing in a variety of cancer types and contexts leading to community-driven improvements and refinements in subsequent applications.

  14. Impact of statistical learning methods on the predictive power of multivariate normal tissue complication probability models

    NARCIS (Netherlands)

    Xu, Cheng-Jian; van der Schaaf, Arjen; Schilstra, Cornelis; Langendijk, Johannes A.; van t Veld, Aart A.

    2012-01-01

    PURPOSE: To study the impact of different statistical learning methods on the prediction performance of multivariate normal tissue complication probability (NTCP) models. METHODS AND MATERIALS: In this study, three learning methods, stepwise selection, least absolute shrinkage and selection operator

  15. A machine learning approach to the accurate prediction of monitor units for a compact proton machine.

    Science.gov (United States)

    Sun, Baozhou; Lam, Dao; Yang, Deshan; Grantham, Kevin; Zhang, Tiezhi; Mutic, Sasa; Zhao, Tianyu

    2018-05-01

    Clinical treatment planning systems for proton therapy currently do not calculate monitor units (MUs) in passive scatter proton therapy due to the complexity of the beam delivery systems. Physical phantom measurements are commonly employed to determine the field-specific output factors (OFs) but are often subject to limited machine time, measurement uncertainties and intensive labor. In this study, a machine learning-based approach was developed to predict output (cGy/MU) and derive MUs, incorporating the dependencies on gantry angle and field size for a single-room proton therapy system. The goal of this study was to develop a secondary check tool for OF measurements and eventually eliminate patient-specific OF measurements. The OFs of 1754 fields previously measured in a water phantom with calibrated ionization chambers and electrometers for patient-specific fields with various range and modulation width combinations for 23 options were included in this study. The training data sets for machine learning models in three different methods (Random Forest, XGBoost and Cubist) included 1431 (~81%) OFs. Ten-fold cross-validation was used to prevent "overfitting" and to validate each model. The remaining 323 (~19%) OFs were used to test the trained models. The difference between the measured and predicted values from machine learning models was analyzed. Model prediction accuracy was also compared with that of the semi-empirical model developed by Kooy (Phys. Med. Biol. 50, 2005). Additionally, gantry angle dependence of OFs was measured for three groups of options categorized on the selection of the second scatters. Field size dependence of OFs was investigated for the measurements with and without patient-specific apertures. All three machine learning methods showed higher accuracy than the semi-empirical model which shows considerably large discrepancy of up to 7.7% for the treatment fields with full range and full modulation width. The Cubist-based solution

  16. Collaborative testing as a learning strategy in nursing education.

    Science.gov (United States)

    Sandahl, Sheryl S

    2010-01-01

    A primary goal of nursing education is to prepare nurses to work collaboratively as members of interprofessional health care teams on behalf of patients. Collaborative testing is a collaborative learning strategy used to foster knowledge development, critical thinking in decision making, and group processing skills. This study incorporated a quasi-experimental design with a comparison group to examine the effect of collaborative testing as a learning strategy on student learning and retention of course content as well as group process skills and student perceptions of their learning and anxiety. The setting was a baccalaureate nursing program; the sample consisted of two groups of senior students enrolled in Medical-Surgical Nursing II. Student learning, as measured by unit examination scores, was greater for students taking examinations collaboratively compared to individually. Retention of course content, as measured by final examination scores, was not greater for students taking examinations collaboratively compared to individually. Student perceptions were overwhelmingly positive, with students reporting increased learning as a result of the collaborative testing experiences. Despite the lack of data to support increased retention, collaborative testing may be a learning strategy worth implementing in nursing education. Students reported more positive interactions and collaboration with their peers, skills required by the professional nurse.

  17. Blind Test of Physics-Based Prediction of Protein Structures

    Science.gov (United States)

    Shell, M. Scott; Ozkan, S. Banu; Voelz, Vincent; Wu, Guohong Albert; Dill, Ken A.

    2009-01-01

    We report here a multiprotein blind test of a computer method to predict native protein structures based solely on an all-atom physics-based force field. We use the AMBER 96 potential function with an implicit (GB/SA) model of solvation, combined with replica-exchange molecular-dynamics simulations. Coarse conformational sampling is performed using the zipping and assembly method (ZAM), an approach that is designed to mimic the putative physical routes of protein folding. ZAM was applied to the folding of six proteins, from 76 to 112 monomers in length, in CASP7, a community-wide blind test of protein structure prediction. Because these predictions have about the same level of accuracy as typical bioinformatics methods, and do not utilize information from databases of known native structures, this work opens up the possibility of predicting the structures of membrane proteins, synthetic peptides, or other foldable polymers, for which there is little prior knowledge of native structures. This approach may also be useful for predicting physical protein folding routes, non-native conformations, and other physical properties from amino acid sequences. PMID:19186130

  18. Simple exercise test for the prediction of relative heat tolerance

    International Nuclear Information System (INIS)

    Kenney, W.L.; Lewis, D.A.; Anderson, R.K.; Kamon, E.

    1986-01-01

    A medical screening exercise test is presented which accurately predicts relative heat tolerance during work in very hot environments. The test consisted of 15-20 min of exercise at a standard absolute intensity of about 600 kcal/hr (140W) with the subject wearing a vapor-barrier suit. Five minutes after the subject exercised, recovery heart rate was measured. When this heart rate is used, a physiological limit (+/- approximately 5 min) can be predicted with 95% confidence for the most intense work-heat conditions found in nuclear power stations. In addition, site health and safety personnel can establish qualification criteria for work on hot jobs, based on the test results. The test as developed can be performed in an office environment with the use of a minimum of equipment by personnel with minimal expertise and training. Total maximal test duration is about 20-25 min per person and only heart rate need be monitored (simple pulse palpation will suffice). Test modality is adaptable to any ergometer, the most readily available and least expensive of which is bench-stepping. It is recommended that this test be available for use for those persons who, based upon routine medical examination or past history, are suspected of being relatively heat intolerant

  19. Application of Machine Learning to Predict Dietary Lapses During Weight Loss.

    Science.gov (United States)

    Goldstein, Stephanie P; Zhang, Fengqing; Thomas, John G; Butryn, Meghan L; Herbert, James D; Forman, Evan M

    2018-05-01

    Individuals who adhere to dietary guidelines provided during weight loss interventions tend to be more successful with weight control. Any deviation from dietary guidelines can be referred to as a "lapse." There is a growing body of research showing that lapses are predictable using a variety of physiological, environmental, and psychological indicators. With recent technological advancements, it may be possible to assess these triggers and predict dietary lapses in real time. The current study sought to use machine learning techniques to predict lapses and evaluate the utility of combining both group- and individual-level data to enhance lapse prediction. The current study trained and tested a machine learning algorithm capable of predicting dietary lapses from a behavioral weight loss program among adults with overweight/obesity (n = 12). Participants were asked to follow a weight control diet for 6 weeks and complete ecological momentary assessment (EMA; repeated brief surveys delivered via smartphone) regarding dietary lapses and relevant triggers. WEKA decision trees were used to predict lapses with an accuracy of 0.72 for the group of participants. However, generalization of the group algorithm to each individual was poor, and as such, group- and individual-level data were combined to improve prediction. The findings suggest that 4 weeks of individual data collection is recommended to attain optimal model performance. The predictive algorithm could be utilized to provide in-the-moment interventions to prevent dietary lapses and therefore enhance weight losses. Furthermore, methods in the current study could be translated to other types of health behavior lapses.

  20. The machine learning in the prediction of elections

    Directory of Open Access Journals (Sweden)

    M.S.C. José A. León-Borges

    2015-05-01

    Full Text Available This research article, presents an analysis and a comparison of three different algorithms: A.- Grouping method K-means, B.-Expectation a convergence criteria, EM and C.- Methodology for classification LAMDA, using two software of classification Weka and SALSA, as an aid for the prediction of future elections in the state of Quintana Roo. When working with electoral data, these are classified in a qualitative and quantitative way, by such virtue at the end of this article you will have the elements necessary to decide, which software, has better performance for such learning of classification. The main reason for the development of this work, is to demonstrate the efficiency of algorithms, with different data types. At the end, it may be decided, the algorithm with the better performance in data management.

  1. Application of Machine-Learning Models to Predict Tacrolimus Stable Dose in Renal Transplant Recipients

    Science.gov (United States)

    Tang, Jie; Liu, Rong; Zhang, Yue-Li; Liu, Mou-Ze; Hu, Yong-Fang; Shao, Ming-Jie; Zhu, Li-Jun; Xin, Hua-Wen; Feng, Gui-Wen; Shang, Wen-Jun; Meng, Xiang-Guang; Zhang, Li-Rong; Ming, Ying-Zi; Zhang, Wei

    2017-02-01

    Tacrolimus has a narrow therapeutic window and considerable variability in clinical use. Our goal was to compare the performance of multiple linear regression (MLR) and eight machine learning techniques in pharmacogenetic algorithm-based prediction of tacrolimus stable dose (TSD) in a large Chinese cohort. A total of 1,045 renal transplant patients were recruited, 80% of which were randomly selected as the “derivation cohort” to develop dose-prediction algorithm, while the remaining 20% constituted the “validation cohort” to test the final selected algorithm. MLR, artificial neural network (ANN), regression tree (RT), multivariate adaptive regression splines (MARS), boosted regression tree (BRT), support vector regression (SVR), random forest regression (RFR), lasso regression (LAR) and Bayesian additive regression trees (BART) were applied and their performances were compared in this work. Among all the machine learning models, RT performed best in both derivation [0.71 (0.67-0.76)] and validation cohorts [0.73 (0.63-0.82)]. In addition, the ideal rate of RT was 4% higher than that of MLR. To our knowledge, this is the first study to use machine learning models to predict TSD, which will further facilitate personalized medicine in tacrolimus administration in the future.

  2. Investigating Mesoscale Convective Systems and their Predictability Using Machine Learning

    Science.gov (United States)

    Daher, H.; Duffy, D.; Bowen, M. K.

    2016-12-01

    A mesoscale convective system (MCS) is a thunderstorm region that lasts several hours long and forms near weather fronts and can often develop into tornadoes. Here we seek to answer the question of whether these tornadoes are "predictable" by looking for a defining characteristic(s) separating MCSs that evolve into tornadoes versus those that do not. Using NASA's Modern Era Retrospective-analysis for Research and Applications 2 reanalysis data (M2R12K), we apply several state of the art machine learning techniques to investigate this question. The spatial region examined in this experiment is Tornado Alley in the United States over the peak tornado months. A database containing select variables from M2R12K is created using PostgreSQL. This database is then analyzed using machine learning methods such as Symbolic Aggregate approXimation (SAX) and DBSCAN (an unsupervised density-based data clustering algorithm). The incentive behind using these methods is to mathematically define a MCS so that association rule mining techniques can be used to uncover some sort of signal or teleconnection that will help us forecast which MCSs will result in tornadoes and therefore give society more time to prepare and in turn reduce casualties and destruction.

  3. Mini-review: Prediction errors, attention and associative learning.

    Science.gov (United States)

    Holland, Peter C; Schiffino, Felipe L

    2016-05-01

    Most modern theories of associative learning emphasize a critical role for prediction error (PE, the difference between received and expected events). One class of theories, exemplified by the Rescorla-Wagner (1972) model, asserts that PE determines the effectiveness of the reinforcer or unconditioned stimulus (US): surprising reinforcers are more effective than expected ones. A second class, represented by the Pearce-Hall (1980) model, argues that PE determines the associability of conditioned stimuli (CSs), the rate at which they may enter into new learning: the surprising delivery or omission of a reinforcer enhances subsequent processing of the CSs that were present when PE was induced. In this mini-review we describe evidence, mostly from our laboratory, for PE-induced changes in the associability of both CSs and USs, and the brain systems involved in the coding, storage and retrieval of these altered associability values. This evidence favors a number of modifications to behavioral models of how PE influences event processing, and suggests the involvement of widespread brain systems in animals' responses to PE. Copyright © 2016 Elsevier Inc. All rights reserved.

  4. Solar Flare Prediction Model with Three Machine-learning Algorithms using Ultraviolet Brightening and Vector Magnetograms

    Science.gov (United States)

    Nishizuka, N.; Sugiura, K.; Kubo, Y.; Den, M.; Watari, S.; Ishii, M.

    2017-02-01

    We developed a flare prediction model using machine learning, which is optimized to predict the maximum class of flares occurring in the following 24 hr. Machine learning is used to devise algorithms that can learn from and make decisions on a huge amount of data. We used solar observation data during the period 2010-2015, such as vector magnetograms, ultraviolet (UV) emission, and soft X-ray emission taken by the Solar Dynamics Observatory and the Geostationary Operational Environmental Satellite. We detected active regions (ARs) from the full-disk magnetogram, from which ˜60 features were extracted with their time differentials, including magnetic neutral lines, the current helicity, the UV brightening, and the flare history. After standardizing the feature database, we fully shuffled and randomly separated it into two for training and testing. To investigate which algorithm is best for flare prediction, we compared three machine-learning algorithms: the support vector machine, k-nearest neighbors (k-NN), and extremely randomized trees. The prediction score, the true skill statistic, was higher than 0.9 with a fully shuffled data set, which is higher than that for human forecasts. It was found that k-NN has the highest performance among the three algorithms. The ranking of the feature importance showed that previous flare activity is most effective, followed by the length of magnetic neutral lines, the unsigned magnetic flux, the area of UV brightening, and the time differentials of features over 24 hr, all of which are strongly correlated with the flux emergence dynamics in an AR.

  5. Solar Flare Prediction Model with Three Machine-learning Algorithms using Ultraviolet Brightening and Vector Magnetograms

    Energy Technology Data Exchange (ETDEWEB)

    Nishizuka, N.; Kubo, Y.; Den, M.; Watari, S.; Ishii, M. [Applied Electromagnetic Research Institute, National Institute of Information and Communications Technology, 4-2-1, Nukui-Kitamachi, Koganei, Tokyo 184-8795 (Japan); Sugiura, K., E-mail: nishizuka.naoto@nict.go.jp [Advanced Speech Translation Research and Development Promotion Center, National Institute of Information and Communications Technology (Japan)

    2017-02-01

    We developed a flare prediction model using machine learning, which is optimized to predict the maximum class of flares occurring in the following 24 hr. Machine learning is used to devise algorithms that can learn from and make decisions on a huge amount of data. We used solar observation data during the period 2010–2015, such as vector magnetograms, ultraviolet (UV) emission, and soft X-ray emission taken by the Solar Dynamics Observatory and the Geostationary Operational Environmental Satellite . We detected active regions (ARs) from the full-disk magnetogram, from which ∼60 features were extracted with their time differentials, including magnetic neutral lines, the current helicity, the UV brightening, and the flare history. After standardizing the feature database, we fully shuffled and randomly separated it into two for training and testing. To investigate which algorithm is best for flare prediction, we compared three machine-learning algorithms: the support vector machine, k-nearest neighbors (k-NN), and extremely randomized trees. The prediction score, the true skill statistic, was higher than 0.9 with a fully shuffled data set, which is higher than that for human forecasts. It was found that k-NN has the highest performance among the three algorithms. The ranking of the feature importance showed that previous flare activity is most effective, followed by the length of magnetic neutral lines, the unsigned magnetic flux, the area of UV brightening, and the time differentials of features over 24 hr, all of which are strongly correlated with the flux emergence dynamics in an AR.

  6. Solar Flare Prediction Model with Three Machine-learning Algorithms using Ultraviolet Brightening and Vector Magnetograms

    International Nuclear Information System (INIS)

    Nishizuka, N.; Kubo, Y.; Den, M.; Watari, S.; Ishii, M.; Sugiura, K.

    2017-01-01

    We developed a flare prediction model using machine learning, which is optimized to predict the maximum class of flares occurring in the following 24 hr. Machine learning is used to devise algorithms that can learn from and make decisions on a huge amount of data. We used solar observation data during the period 2010–2015, such as vector magnetograms, ultraviolet (UV) emission, and soft X-ray emission taken by the Solar Dynamics Observatory and the Geostationary Operational Environmental Satellite . We detected active regions (ARs) from the full-disk magnetogram, from which ∼60 features were extracted with their time differentials, including magnetic neutral lines, the current helicity, the UV brightening, and the flare history. After standardizing the feature database, we fully shuffled and randomly separated it into two for training and testing. To investigate which algorithm is best for flare prediction, we compared three machine-learning algorithms: the support vector machine, k-nearest neighbors (k-NN), and extremely randomized trees. The prediction score, the true skill statistic, was higher than 0.9 with a fully shuffled data set, which is higher than that for human forecasts. It was found that k-NN has the highest performance among the three algorithms. The ranking of the feature importance showed that previous flare activity is most effective, followed by the length of magnetic neutral lines, the unsigned magnetic flux, the area of UV brightening, and the time differentials of features over 24 hr, all of which are strongly correlated with the flux emergence dynamics in an AR.

  7. Test-retest reliability and predictive validity of the Implicit Association Test in children.

    Science.gov (United States)

    Rae, James R; Olson, Kristina R

    2018-02-01

    The Implicit Association Test (IAT) is increasingly used in developmental research despite minimal evidence of whether children's IAT scores are reliable across time or predictive of behavior. When test-retest reliability and predictive validity have been assessed, the results have been mixed, and because these studies have differed on many factors simultaneously (lag-time between testing administrations, domain, etc.), it is difficult to discern what factors may explain variability in existing test-retest reliability and predictive validity estimates. Across five studies (total N = 519; ages 6- to 11-years-old), we manipulated two factors that have varied in previous developmental research-lag-time and domain. An internal meta-analysis of these studies revealed that, across three different methods of analyzing the data, mean test-retest (rs of .48, .38, and .34) and predictive validity (rs of .46, .20, and .10) effect sizes were significantly greater than zero. While lag-time did not moderate the magnitude of test-retest coefficients, whether we observed domain differences in test-retest reliability and predictive validity estimates was contingent on other factors, such as how we scored the IAT or whether we included estimates from a unique sample (i.e., a sample containing gender typical and gender diverse children). Recommendations are made for developmental researchers that utilize the IAT in their research. (PsycINFO Database Record (c) 2018 APA, all rights reserved).

  8. Testing the Predictive Power of Coulomb Stress on Aftershock Sequences

    Science.gov (United States)

    Woessner, J.; Lombardi, A.; Werner, M. J.; Marzocchi, W.

    2009-12-01

    Empirical and statistical models of clustered seismicity are usually strongly stochastic and perceived to be uninformative in their forecasts, since only marginal distributions are used, such as the Omori-Utsu and Gutenberg-Richter laws. In contrast, so-called physics-based aftershock models, based on seismic rate changes calculated from Coulomb stress changes and rate-and-state friction, make more specific predictions: anisotropic stress shadows and multiplicative rate changes. We test the predictive power of models based on Coulomb stress changes against statistical models, including the popular Short Term Earthquake Probabilities and Epidemic-Type Aftershock Sequences models: We score and compare retrospective forecasts on the aftershock sequences of the 1992 Landers, USA, the 1997 Colfiorito, Italy, and the 2008 Selfoss, Iceland, earthquakes. To quantify predictability, we use likelihood-based metrics that test the consistency of the forecasts with the data, including modified and existing tests used in prospective forecast experiments within the Collaboratory for the Study of Earthquake Predictability (CSEP). Our results indicate that a statistical model performs best. Moreover, two Coulomb model classes seem unable to compete: Models based on deterministic Coulomb stress changes calculated from a given fault-slip model, and those based on fixed receiver faults. One model of Coulomb stress changes does perform well and sometimes outperforms the statistical models, but its predictive information is diluted, because of uncertainties included in the fault-slip model. Our results suggest that models based on Coulomb stress changes need to incorporate stochastic features that represent model and data uncertainty.

  9. Individual differences in spatial configuration learning predict the occurrence of intrusive memories.

    Science.gov (United States)

    Meyer, Thomas; Smeets, Tom; Giesbrecht, Timo; Quaedflieg, Conny W E M; Girardelli, Marta M; Mackay, Georgina R N; Merckelbach, Harald

    2013-03-01

    The dual-representation model of posttraumatic stress disorder (PTSD; Brewin, Gregory, Lipton, & Burgess, Psychological Review, 117, 210-232 2010) argues that intrusions occur when people fail to construct context-based representations during adverse experiences. The present study tested a specific prediction flowing from this model. In particular, we investigated whether the efficiency of temporal-lobe-based spatial configuration learning would account for individual differences in intrusive experiences and physiological reactivity in the laboratory. Participants (N = 82) completed the contextual cuing paradigm, which assesses spatial configuration learning that is believed to depend on associative encoding in the parahippocampus. They were then shown a trauma film. Afterward, startle responses were quantified during presentation of trauma reminder pictures versus unrelated neutral and emotional pictures. PTSD symptoms were recorded in the week following participation. Better configuration learning performance was associated with fewer perceptual intrusions, r = -.33, p .46) and had no direct effect on intrusion-related distress and overall PTSD symptoms, rs > -.12, ps > .29. However, configuration learning performance tended to be associated with reduced physiological responses to unrelated negative images, r = -.20, p = .07. Thus, while spatial configuration learning appears to be unrelated to affective responding to trauma reminders, our overall findings support the idea that the context-based memory system helps to reduce intrusions.

  10. INSIGHTS FROM MACHINE-LEARNED DIET SUCCESS PREDICTION.

    Science.gov (United States)

    Weber, Ingmar; Achananuparp, Palakorn

    2016-01-01

    To support people trying to lose weight and stay healthy, more and more fitness apps have sprung up including the ability to track both calories intake and expenditure. Users of such apps are part of a wider "quantified self" movement and many opt-in to publicly share their logged data. In this paper, we use public food diaries of more than 4,000 long-term active MyFitnessPal users to study the characteristics of a (un-)successful diet. Concretely, we train a machine learning model to predict repeatedly being over or under self-set daily calories goals and then look at which features contribute to the model's prediction. Our findings include both expected results, such as the token "mcdonalds" or the category "dessert" being indicative for being over the calories goal, but also less obvious ones such as the difference between pork and poultry concerning dieting success, or the use of the "quick added calories" functionality being indicative of over-shooting calorie-wise. This study also hints at the feasibility of using such data for more in-depth data mining, e.g., looking at the interaction between consumed foods such as mixing protein- and carbohydrate-rich foods. To the best of our knowledge, this is the first systematic study of public food diaries.

  11. Paired-Associate and Feedback-Based Weather Prediction Tasks Support Multiple Category Learning Systems

    OpenAIRE

    Li, Kaiyun; Fu, Qiufang; Sun, Xunwei; Zhou, Xiaoyan; Fu, Xiaolan

    2016-01-01

    It remains unclear whether probabilistic category learning in the feedback-based weather prediction task (FB-WPT) can be mediated by a non-declarative or procedural learning system. To address this issue, we compared the effects of training time and verbal working memory, which influence the declarative learning system but not the non-declarative learning system, in the FB and paired-associate (PA) WPTs, as the PA task recruits a declarative learning system. The results of Experiment 1 showed...

  12. Causal beliefs about depression in different cultural groups—what do cognitive psychological theories of causal learning and reasoning predict?

    OpenAIRE

    Hagmayer, York; Engelmann, Neele

    2014-01-01

    Cognitive psychological research focuses on causal learning and reasoning while cognitive anthropological and social science research tend to focus on systems of beliefs. Our aim was to explore how these two types of research can inform each other. Cognitive psychological theories (causal model theory and causal Bayes nets) were used to derive predictions for systems of causal beliefs. These predictions were then applied to lay theories of depression as a specific test case. A systematic lite...

  13. Machine learning in updating predictive models of planning and scheduling transportation projects

    Science.gov (United States)

    1997-01-01

    A method combining machine learning and regression analysis to automatically and intelligently update predictive models used in the Kansas Department of Transportations (KDOTs) internal management system is presented. The predictive models used...

  14. The conditions that promote fear learning: prediction error and Pavlovian fear conditioning.

    Science.gov (United States)

    Li, Susan Shi Yuan; McNally, Gavan P

    2014-02-01

    A key insight of associative learning theory is that learning depends on the actions of prediction error: a discrepancy between the actual and expected outcomes of a conditioning trial. When positive, such error causes increments in associative strength and, when negative, such error causes decrements in associative strength. Prediction error can act directly on fear learning by determining the effectiveness of the aversive unconditioned stimulus or indirectly by determining the effectiveness, or associability, of the conditioned stimulus. Evidence from a variety of experimental preparations in human and non-human animals suggest that discrete neural circuits code for these actions of prediction error during fear learning. Here we review the circuits and brain regions contributing to the neural coding of prediction error during fear learning and highlight areas of research (safety learning, extinction, and reconsolidation) that may profit from this approach to understanding learning. Crown Copyright © 2013. Published by Elsevier Inc. All rights reserved.

  15. Controlling misses and false alarms in a machine learning framework for predicting uniformity of printed pages

    Science.gov (United States)

    Nguyen, Minh Q.; Allebach, Jan P.

    2015-01-01

    In our previous work1 , we presented a block-based technique to analyze printed page uniformity both visually and metrically. The features learned from the models were then employed in a Support Vector Machine (SVM) framework to classify the pages into one of the two categories of acceptable and unacceptable quality. In this paper, we introduce a set of tools for machine learning in the assessment of printed page uniformity. This work is primarily targeted to the printing industry, specifically the ubiquitous laser, electrophotographic printer. We use features that are well-correlated with the rankings of expert observers to develop a novel machine learning framework that allows one to achieve the minimum "false alarm" rate, subject to a chosen "miss" rate. Surprisingly, most of the research that has been conducted on machine learning does not consider this framework. During the process of developing a new product, test engineers will print hundreds of test pages, which can be scanned and then analyzed by an autonomous algorithm. Among these pages, most may be of acceptable quality. The objective is to find the ones that are not. These will provide critically important information to systems designers, regarding issues that need to be addressed in improving the printer design. A "miss" is defined to be a page that is not of acceptable quality to an expert observer that the prediction algorithm declares to be a "pass". Misses are a serious problem, since they represent problems that will not be seen by the systems designers. On the other hand, "false alarms" correspond to pages that an expert observer would declare to be of acceptable quality, but which are flagged by the prediction algorithm as "fails". In a typical printer testing and development scenario, such pages would be examined by an expert, and found to be of acceptable quality after all. "False alarm" pages result in extra pages to be examined by expert observers, which increases labor cost. But "false

  16. Knowledge Prediction of Different Students’ Categories Trough an Intelligent Testing

    Directory of Open Access Journals (Sweden)

    Irina Zheliazkova

    2015-02-01

    Full Text Available Student’s modelling, prediction, and grouping have remained open research issues in the multi-disciplinary area of educational data mining. The purpose of this study is to predict the correct knowledge of different categories of tested students: good, very good, and all. The experimental data set was gathered from an intelligent post-test performance containing student’s correct, missing, and wrong knowledge, time undertaken, and final mark. The proposed procedure applies consequently correlation analysis, simple and multiple liner regression using a power specialized tool for programming by the teacher. The finding is that the accuracy of the procedure is satisfactory for the three students’ categories. The experiment also confirms some findings of other researchers and previous authors’ team studies.

  17. Psychodynamic theory and counseling in predictive testing for Huntington's disease.

    Science.gov (United States)

    Tassicker, Roslyn J

    2005-04-01

    This paper revisits psychodynamic theory, which can be applied in predictive testing counseling for Huntington's Disease (HD). Psychodynamic theory has developed from the work of Freud and places importance on early parent-child experiences. The nature of these relationships, or attachments are reflected in adult expectations and relationships. Two significant concepts, identification and fear of abandonment, have been developed and expounded by the psychodynamic theorist, Melanie Klein. The processes of identification and fear of abandonment can become evident in predictive testing counseling and are colored by the client's experience of growing up with a parent affected by Huntington's Disease. In reflecting on family-of-origin experiences, clients can also express implied expectations of the future, and future relationships. Case examples are given to illustrate the dynamic processes of identification and fear of abandonment which may present in the clinical setting. Counselor recognition of these processes can illuminate and inform counseling practice.

  18. Testing and Life Prediction for Composite Rotor Hub Flexbeams

    Science.gov (United States)

    Murri, Gretchen B.

    2004-01-01

    A summary of several studies of delamination in tapered composite laminates with internal ply-drops is presented. Initial studies used 2D FE models to calculate interlaminar stresses at the ply-ending locations in linear tapered laminates under tension loading. Strain energy release rates for delamination in these laminates indicated that delamination would likely start at the juncture of the tapered and thin regions and grow unstably in both directions. Tests of glass/epoxy and graphite/epoxy linear tapered laminates under axial tension delaminated as predicted. Nonlinear tapered specimens were cut from a full-size helicopter rotor hub and were tested under combined constant axial tension and cyclic transverse bending loading to simulate the loading experienced by a rotorhub flexbeam in flight. For all the tested specimens, delamination began at the tip of the outermost dropped ply group and grew first toward the tapered region. A 2D FE model was created that duplicated the test flexbeam layup, geometry, and loading. Surface strains calculated by the model agreed very closely with the measured surface strains in the specimens. The delamination patterns observed in the tests were simulated in the model by releasing pairs of MPCs along those interfaces. Strain energy release rates associated with the delamination growth were calculated for several configurations and using two different FE analysis codes. Calculations from the codes agreed very closely. The strain energy release rate results were used with material characterization data to predict fatigue delamination onset lives for nonlinear tapered flexbeams with two different ply-dropping schemes. The predicted curves agreed well with the test data for each case studied.

  19. Generating pseudo test collections for learning to rank scientific articles

    NARCIS (Netherlands)

    Berendsen, R.; Tsagkias, M.; de Rijke, M.; Meij, E.

    2012-01-01

    Pseudo test collections are automatically generated to provide training material for learning to rank methods. We propose a method for generating pseudo test collections in the domain of digital libraries, where data is relatively sparse, but comes with rich annotations. Our intuition is that

  20. Learned Helplessness, Test Anxiety, and Academic Achievement: A Longitudinal Analysis.

    Science.gov (United States)

    Fincham, Frank D.; And Others

    1989-01-01

    Examines the stability of individual differences in test anxiety and learned helplessness of 82 children in third grade and later in fifth grade. Results indicate that teacher reports of helplessness had the strongest and most consistent relation to concurrent achievement and to achievement test scores two years later. (RJC)

  1. A Serious Flaw in the Collegiate Learning Assessment [CLA] Test

    Directory of Open Access Journals (Sweden)

    Kevin Possin

    2013-09-01

    Full Text Available The Collegiate Learning Assessment Test (CLA has become popular and highly recommended, praised for its reliability and validity. I argue that while the CLA may be a commendable test for measuring critical-thinking, problem-solving, and logical-reasoning skills, those who are scoring students’ answers to the test’s questions are rendering the CLA invalid.

  2. Powerful Tests for Multi-Marker Association Analysis Using Ensemble Learning.

    Directory of Open Access Journals (Sweden)

    Badri Padhukasahasram

    Full Text Available Multi-marker approaches have received a lot of attention recently in genome wide association studies and can enhance power to detect new associations under certain conditions. Gene-, gene-set- and pathway-based association tests are increasingly being viewed as useful supplements to the more widely used single marker association analysis which have successfully uncovered numerous disease variants. A major drawback of single-marker based methods is that they do not look at the joint effects of multiple genetic variants which individually may have weak or moderate signals. Here, we describe novel tests for multi-marker association analyses that are based on phenotype predictions obtained from machine learning algorithms. Instead of assuming a linear or logistic regression model, we propose the use of ensembles of diverse machine learning algorithms for prediction. We show that phenotype predictions obtained from ensemble learning algorithms provide a new framework for multi-marker association analysis. They can be used for constructing tests for the joint association of multiple variants, adjusting for covariates and testing for the presence of interactions. To demonstrate the power and utility of this new approach, we first apply our method to simulated SNP datasets. We show that the proposed method has the correct Type-1 error rates and can be considerably more powerful than alternative approaches in some situations. Then, we apply our method to previously studied asthma-related genes in 2 independent asthma cohorts to conduct association tests.

  3. A Hybrid Supervised/Unsupervised Machine Learning Approach to Solar Flare Prediction

    Science.gov (United States)

    Benvenuto, Federico; Piana, Michele; Campi, Cristina; Massone, Anna Maria

    2018-01-01

    This paper introduces a novel method for flare forecasting, combining prediction accuracy with the ability to identify the most relevant predictive variables. This result is obtained by means of a two-step approach: first, a supervised regularization method for regression, namely, LASSO is applied, where a sparsity-enhancing penalty term allows the identification of the significance with which each data feature contributes to the prediction; then, an unsupervised fuzzy clustering technique for classification, namely, Fuzzy C-Means, is applied, where the regression outcome is partitioned through the minimization of a cost function and without focusing on the optimization of a specific skill score. This approach is therefore hybrid, since it combines supervised and unsupervised learning; realizes classification in an automatic, skill-score-independent way; and provides effective prediction performances even in the case of imbalanced data sets. Its prediction power is verified against NOAA Space Weather Prediction Center data, using as a test set, data in the range between 1996 August and 2010 December and as training set, data in the range between 1988 December and 1996 June. To validate the method, we computed several skill scores typically utilized in flare prediction and compared the values provided by the hybrid approach with the ones provided by several standard (non-hybrid) machine learning methods. The results showed that the hybrid approach performs classification better than all other supervised methods and with an effectiveness comparable to the one of clustering methods; but, in addition, it provides a reliable ranking of the weights with which the data properties contribute to the forecast.

  4. Grammar predicts procedural learning and consolidation deficits in children with Specific Language Impairment.

    Science.gov (United States)

    Hedenius, Martina; Persson, Jonas; Tremblay, Antoine; Adi-Japha, Esther; Veríssimo, João; Dye, Cristina D; Alm, Per; Jennische, Margareta; Bruce Tomblin, J; Ullman, Michael T

    2011-01-01

    The Procedural Deficit Hypothesis (PDH) posits that Specific Language Impairment (SLI) can be largely explained by abnormalities of brain structures that subserve procedural memory. The PDH predicts impairments of procedural memory itself, and that such impairments underlie the grammatical deficits observed in the disorder. Previous studies have indeed reported procedural learning impairments in SLI, and have found that these are associated with grammatical difficulties. The present study extends this research by examining consolidation and longer-term procedural sequence learning in children with SLI. The Alternating Serial Reaction Time (ASRT) task was given to children with SLI and typically developing (TD) children in an initial learning session and an average of three days later to test for consolidation and longer-term learning. Although both groups showed evidence of initial sequence learning, only the TD children showed clear signs of consolidation, even though the two groups did not differ in longer-term learning. When the children were re-categorized on the basis of grammar deficits rather than broader language deficits, a clearer pattern emerged. Whereas both the grammar impaired and normal grammar groups showed evidence of initial sequence learning, only those with normal grammar showed consolidation and longer-term learning. Indeed, the grammar-impaired group appeared to lose any sequence knowledge gained during the initial testing session. These findings held even when controlling for vocabulary or a broad non-grammatical language measure, neither of which were associated with procedural memory. When grammar was examined as a continuous variable over all children, the same relationships between procedural memory and grammar, but not vocabulary or the broader language measure, were observed. Overall, the findings support and further specify the PDH. They suggest that consolidation and longer-term procedural learning are impaired in SLI, but that these

  5. Grammar Predicts Procedural Learning and Consolidation Deficits in Children with Specific Language Impairment

    Science.gov (United States)

    Hedenius, Martina; Persson, Jonas; Tremblay, Antoine; Adi-Japha, Esther; Veríssimo, João; Dye, Cristina D.; Alm, Per; Jennische, Margareta; Tomblin, J. Bruce; Ullman, Michael T.

    2011-01-01

    The Procedural Deficit Hypothesis (PDH) posits that Specific Language Impairment (SLI) can be largely explained by abnormalities of brain structures that subserve procedural memory. The PDH predicts impairments of procedural memory itself, and that such impairments underlie the grammatical deficits observed in the disorder. Previous studies have indeed reported procedural learning impairments in SLI, and have found that these are associated with grammatical difficulties. The present study extends this research by examining the consolidation and longer-term procedural sequence learning in children with SLI. The Alternating Serial Reaction Time (ASRT) task was given to children with SLI and typically-developing (TD) children in an initial learning session and an average of three days later to test for consolidation and longer-term learning. Although both groups showed evidence of initial sequence learning, only the TD children showed clear signs of consolidation, even though the two groups did not differ in longer-term learning. When the children were re-categorized on the basis of grammar deficits rather than broader language deficits, a clearer pattern emerged. Whereas both the grammar impaired and normal grammar groups showed evidence of initial sequence learning, only those with normal grammar showed consolidation and longer-term learning. Indeed, the grammar-impaired group appeared to lose any sequence knowledge gained during the initial testing session. These findings held even when controlling for vocabulary or a broad non-grammatical language measure, neither of which were associated with procedural memory. When grammar was examined as a continuous variable over all children, the same relationships between procedural memory and grammar, but not vocabulary or the broader language measure, were observed. Overall, the findings support and further specify the PDH. They suggest that consolidation and longer-term procedural learning are impaired in SLI, but that

  6. Predicting the Failure of Dental Implants Using Supervised Learning Techniques

    Directory of Open Access Journals (Sweden)

    Chia-Hui Liu

    2018-05-01

    Full Text Available Prosthodontic treatment has been a crucial part of dental treatment for patients with full mouth rehabilitation. Dental implant surgeries that replace conventional dentures using titanium fixtures have become the top choice. However, because of the wide-ranging scope of implant surgeries, patients’ body conditions, surgeons’ experience, and the choice of implant system should be considered during treatment. The higher price charged by dental implant treatments compared to conventional dentures has led to a rush among medical staff; therefore, the future impact of surgeries has not been analyzed in detail, resulting in medial disputes. Previous literature on the success factors of dental implants is mainly focused on single factors such as patients’ systemic diseases, operation methods, or prosthesis types for statistical correlation significance analysis. This study developed a prediction model for providing an early warning mechanism to reduce the chances of dental implant failure. We collected the clinical data of patients who received artificial dental implants at the case hospital for a total of 8 categories and 20 variables. Supervised learning techniques such as decision tree (DT, support vector machines, logistic regressions, and classifier ensembles (i.e., Bagging and AdaBoost were used to analyze the prediction of the failure of dental implants. The results show that DT with both Bagging and Adaboost techniques possesses the highest prediction performance for the failure of dental implant (area under the receiver operating characteristic curve, AUC: 0.741; the analysis also revealed that the implant systems affect dental implant failure. The model can help clinical surgeons to reduce medical failures by choosing the optimal implant system and prosthodontics treatments for their patients.

  7. Multimodal manifold-regularized transfer learning for MCI conversion prediction.

    Science.gov (United States)

    Cheng, Bo; Liu, Mingxia; Suk, Heung-Il; Shen, Dinggang; Zhang, Daoqiang

    2015-12-01

    As the early stage of Alzheimer's disease (AD), mild cognitive impairment (MCI) has high chance to convert to AD. Effective prediction of such conversion from MCI to AD is of great importance for early diagnosis of AD and also for evaluating AD risk pre-symptomatically. Unlike most previous methods that used only the samples from a target domain to train a classifier, in this paper, we propose a novel multimodal manifold-regularized transfer learning (M2TL) method that jointly utilizes samples from another domain (e.g., AD vs. normal controls (NC)) as well as unlabeled samples to boost the performance of the MCI conversion prediction. Specifically, the proposed M2TL method includes two key components. The first one is a kernel-based maximum mean discrepancy criterion, which helps eliminate the potential negative effect induced by the distributional difference between the auxiliary domain (i.e., AD and NC) and the target domain (i.e., MCI converters (MCI-C) and MCI non-converters (MCI-NC)). The second one is a semi-supervised multimodal manifold-regularized least squares classification method, where the target-domain samples, the auxiliary-domain samples, and the unlabeled samples can be jointly used for training our classifier. Furthermore, with the integration of a group sparsity constraint into our objective function, the proposed M2TL has a capability of selecting the informative samples to build a robust classifier. Experimental results on the Alzheimer's Disease Neuroimaging Initiative (ADNI) database validate the effectiveness of the proposed method by significantly improving the classification accuracy of 80.1 % for MCI conversion prediction, and also outperforming the state-of-the-art methods.

  8. Plant MicroRNA Prediction by Supervised Machine Learning Using C5.0 Decision Trees

    Directory of Open Access Journals (Sweden)

    Philip H. Williams

    2012-01-01

    Full Text Available MicroRNAs (miRNAs are nonprotein coding RNAs between 20 and 22 nucleotides long that attenuate protein production. Different types of sequence data are being investigated for novel miRNAs, including genomic and transcriptomic sequences. A variety of machine learning methods have successfully predicted miRNA precursors, mature miRNAs, and other nonprotein coding sequences. MirTools, mirDeep2, and miRanalyzer require “read count” to be included with the input sequences, which restricts their use to deep-sequencing data. Our aim was to train a predictor using a cross-section of different species to accurately predict miRNAs outside the training set. We wanted a system that did not require read-count for prediction and could therefore be applied to short sequences extracted from genomic, EST, or RNA-seq sources. A miRNA-predictive decision-tree model has been developed by supervised machine learning. It only requires that the corresponding genome or transcriptome is available within a sequence window that includes the precursor candidate so that the required sequence features can be collected. Some of the most critical features for training the predictor are the miRNA:miRNA∗ duplex energy and the number of mismatches in the duplex. We present a cross-species plant miRNA predictor with 84.08% sensitivity and 98.53% specificity based on rigorous testing by leave-one-out validation.

  9. Plant MicroRNA Prediction by Supervised Machine Learning Using C5.0 Decision Trees.

    Science.gov (United States)

    Williams, Philip H; Eyles, Rod; Weiller, Georg

    2012-01-01

    MicroRNAs (miRNAs) are nonprotein coding RNAs between 20 and 22 nucleotides long that attenuate protein production. Different types of sequence data are being investigated for novel miRNAs, including genomic and transcriptomic sequences. A variety of machine learning methods have successfully predicted miRNA precursors, mature miRNAs, and other nonprotein coding sequences. MirTools, mirDeep2, and miRanalyzer require "read count" to be included with the input sequences, which restricts their use to deep-sequencing data. Our aim was to train a predictor using a cross-section of different species to accurately predict miRNAs outside the training set. We wanted a system that did not require read-count for prediction and could therefore be applied to short sequences extracted from genomic, EST, or RNA-seq sources. A miRNA-predictive decision-tree model has been developed by supervised machine learning. It only requires that the corresponding genome or transcriptome is available within a sequence window that includes the precursor candidate so that the required sequence features can be collected. Some of the most critical features for training the predictor are the miRNA:miRNA(∗) duplex energy and the number of mismatches in the duplex. We present a cross-species plant miRNA predictor with 84.08% sensitivity and 98.53% specificity based on rigorous testing by leave-one-out validation.

  10. Multi-Label Learning via Random Label Selection for Protein Subcellular Multi-Locations Prediction.

    Science.gov (United States)

    Wang, Xiao; Li, Guo-Zheng

    2013-03-12

    Prediction of protein subcellular localization is an important but challenging problem, particularly when proteins may simultaneously exist at, or move between, two or more different subcellular location sites. Most of the existing protein subcellular localization methods are only used to deal with the single-location proteins. In the past few years, only a few methods have been proposed to tackle proteins with multiple locations. However, they only adopt a simple strategy, that is, transforming the multi-location proteins to multiple proteins with single location, which doesn't take correlations among different subcellular locations into account. In this paper, a novel method named RALS (multi-label learning via RAndom Label Selection), is proposed to learn from multi-location proteins in an effective and efficient way. Through five-fold cross validation test on a benchmark dataset, we demonstrate our proposed method with consideration of label correlations obviously outperforms the baseline BR method without consideration of label correlations, indicating correlations among different subcellular locations really exist and contribute to improvement of prediction performance. Experimental results on two benchmark datasets also show that our proposed methods achieve significantly higher performance than some other state-of-the-art methods in predicting subcellular multi-locations of proteins. The prediction web server is available at http://levis.tongji.edu.cn:8080/bioinfo/MLPred-Euk/ for the public usage.

  11. Daily sea level prediction at Chiayi coast, Taiwan using extreme learning machine and relevance vector machine

    Science.gov (United States)

    Imani, Moslem; Kao, Huan-Chin; Lan, Wen-Hau; Kuo, Chung-Yen

    2018-02-01

    The analysis and the prediction of sea level fluctuations are core requirements of marine meteorology and operational oceanography. Estimates of sea level with hours-to-days warning times are especially important for low-lying regions and coastal zone management. The primary purpose of this study is to examine the applicability and capability of extreme learning machine (ELM) and relevance vector machine (RVM) models for predicting sea level variations and compare their performances with powerful machine learning methods, namely, support vector machine (SVM) and radial basis function (RBF) models. The input dataset from the period of January 2004 to May 2011 used in the study was obtained from the Dongshi tide gauge station in Chiayi, Taiwan. Results showed that the ELM and RVM models outperformed the other methods. The performance of the RVM approach was superior in predicting the daily sea level time series given the minimum root mean square error of 34.73 mm and the maximum determination coefficient of 0.93 (R2) during the testing periods. Furthermore, the obtained results were in close agreement with the original tide-gauge data, which indicates that RVM approach is a promising alternative method for time series prediction and could be successfully used for daily sea level forecasts.

  12. e-Bitter: Bitterant Prediction by the Consensus Voting From the Machine-Learning Methods.

    Science.gov (United States)

    Zheng, Suqing; Jiang, Mengying; Zhao, Chengwei; Zhu, Rui; Hu, Zhicheng; Xu, Yong; Lin, Fu

    2018-01-01

    In-silico bitterant prediction received the considerable attention due to the expensive and laborious experimental-screening of the bitterant. In this work, we collect the fully experimental dataset containing 707 bitterants and 592 non-bitterants, which is distinct from the fully or partially hypothetical non-bitterant dataset used in the previous works. Based on this experimental dataset, we harness the consensus votes from the multiple machine-learning methods (e.g., deep learning etc.) combined with the molecular fingerprint to build the bitter/bitterless classification models with five-fold cross-validation, which are further inspected by the Y-randomization test and applicability domain analysis. One of the best consensus models affords the accuracy, precision, specificity, sensitivity, F1-score, and Matthews correlation coefficient (MCC) of 0.929, 0.918, 0.898, 0.954, 0.936, and 0.856 respectively on our test set. For the automatic prediction of bitterant, a graphic program "e-Bitter" is developed for the convenience of users via the simple mouse click. To our best knowledge, it is for the first time to adopt the consensus model for the bitterant prediction and develop the first free stand-alone software for the experimental food scientist.

  13. e-Bitter: Bitterant Prediction by the Consensus Voting From the Machine-Learning Methods

    Directory of Open Access Journals (Sweden)

    Suqing Zheng

    2018-03-01

    Full Text Available In-silico bitterant prediction received the considerable attention due to the expensive and laborious experimental-screening of the bitterant. In this work, we collect the fully experimental dataset containing 707 bitterants and 592 non-bitterants, which is distinct from the fully or partially hypothetical non-bitterant dataset used in the previous works. Based on this experimental dataset, we harness the consensus votes from the multiple machine-learning methods (e.g., deep learning etc. combined with the molecular fingerprint to build the bitter/bitterless classification models with five-fold cross-validation, which are further inspected by the Y-randomization test and applicability domain analysis. One of the best consensus models affords the accuracy, precision, specificity, sensitivity, F1-score, and Matthews correlation coefficient (MCC of 0.929, 0.918, 0.898, 0.954, 0.936, and 0.856 respectively on our test set. For the automatic prediction of bitterant, a graphic program “e-Bitter” is developed for the convenience of users via the simple mouse click. To our best knowledge, it is for the first time to adopt the consensus model for the bitterant prediction and develop the first free stand-alone software for the experimental food scientist.

  14. e-Bitter: Bitterant Prediction by the Consensus Voting From the Machine-learning Methods

    Science.gov (United States)

    Zheng, Suqing; Jiang, Mengying; Zhao, Chengwei; Zhu, Rui; Hu, Zhicheng; Xu, Yong; Lin, Fu

    2018-03-01

    In-silico bitterant prediction received the considerable attention due to the expensive and laborious experimental-screening of the bitterant. In this work, we collect the fully experimental dataset containing 707 bitterants and 592 non-bitterants, which is distinct from the fully or partially hypothetical non-bitterant dataset used in the previous works. Based on this experimental dataset, we harness the consensus votes from the multiple machine-learning methods (e.g., deep learning etc.) combined with the molecular fingerprint to build the bitter/bitterless classification models with five-fold cross-validation, which are further inspected by the Y-randomization test and applicability domain analysis. One of the best consensus models affords the accuracy, precision, specificity, sensitivity, F1-score, and Matthews correlation coefficient (MCC) of 0.929, 0.918, 0.898, 0.954, 0.936, and 0.856 respectively on our test set. For the automatic prediction of bitterant, a graphic program “e-Bitter” is developed for the convenience of users via the simple mouse click. To our best knowledge, it is for the first time to adopt the consensus model for the bitterant prediction and develop the first free stand-alone software for the experimental food scientist.

  15. Stress enhances the consolidation of extinction memory in a predictive learning task

    Directory of Open Access Journals (Sweden)

    Tanja eHamacher-Dang

    2013-08-01

    Full Text Available Extinction is not always permanent, as indicated by several types of recovery effects, such as the renewal effect, which may occur after a context change and points towards the importance of contextual cues. Strengthening the retrieval of extinction memory is a crucial aim of extinction-based psychotherapeutic treatments of anxiety disorders to prevent relapse. Stress is known to modulate learning and memory, with mostly enhancing effects on memory consolidation. However, whether such a consolidation-enhancing effect of acute stress can also be found for extinction memory has not yet been examined in humans. In this study, we investigated the effect of stress after extinction learning on the retrieval of extinction memory in a predictive learning renewal paradigm. Participants took the part of being the doctor of a fictitious patient and learned to predict whether certain food stimuli were associated with ‘stomach trouble’ in two different restaurants (contexts. On the first day, critical stimuli were associated with stomach trouble in context A (acquisition phase. On the second day, these associations were extinguished in context B. Directly after extinction, participants were either exposed to a stressor (socially evaluated cold pressor test; n = 22 or a control condition (n = 24. On the third day, we tested retrieval of critical associations in contexts A and B. Participants exposed to stress after extinction exhibited a reduced recovery of responding at test in context B, suggesting that stress may context-dependently enhance the consolidation of extinction memory. Furthermore, the increase in cortisol in response to the stressor was negatively correlated with the recovery of responding in context A. Our findings suggest that in parallel to the known effects of stress on the consolidation of episodic memory, stress also enhances the consolidation of extinction memory, which might be relevant for potential applications in extinction

  16. Predictive gene testing for Huntington disease and other neurodegenerative disorders.

    Science.gov (United States)

    Wedderburn, S; Panegyres, P K; Andrew, S; Goldblatt, J; Liebeck, T; McGrath, F; Wiltshire, M; Pestell, C; Lee, J; Beilby, J

    2013-12-01

    Controversies exist around predictive testing (PT) programmes in neurodegenerative disorders. This study sets out to answer the following questions relating to Huntington disease (HD) and other neurodegenerative disorders: differences between these patients in their PT journeys, why and when individuals withdraw from PT, and decision-making processes regarding reproductive genetic testing. A case series analysis of patients having PT from the multidisciplinary Western Australian centre for PT over the past 20 years was performed using internationally recognised guidelines for predictive gene testing in neurodegenerative disorders. Of 740 at-risk patients, 518 applied for PT: 466 at risk of HD, 52 at risk of other neurodegenerative disorders - spinocerebellar ataxias, hereditary prion disease and familial Alzheimer disease. Thirteen percent withdrew from PT - 80.32% of withdrawals occurred during counselling stages. Major withdrawal reasons related to timing in the patients' lives or unknown as the patient did not disclose the reason. Thirty-eight HD individuals had reproductive genetic testing: 34 initiated prenatal testing (of which eight withdrew from the process) and four initiated pre-implantation genetic diagnosis. There was no recorded or other evidence of major psychological reactions or suicides during PT. People withdrew from PT in relation to life stages and reasons that are unknown. Our findings emphasise the importance of: (i) adherence to internationally recommended guidelines for PT; (ii) the role of the multidisciplinary team in risk minimisation; and (iii) patient selection. © 2013 The Authors; Internal Medicine Journal © 2013 Royal Australasian College of Physicians.

  17. Predicting Freshman Grade Point Average From College Admissions Test Scores and State High School Test Scores

    Directory of Open Access Journals (Sweden)

    Daniel Koretz

    2016-09-01

    Full Text Available The current focus on assessing “college and career readiness” raises an empirical question: How do high school tests compare with college admissions tests in predicting performance in college? We explored this using data from the City University of New York and public colleges in Kentucky. These two systems differ in the choice of college admissions test, the stakes for students on the high school test, and demographics. We predicted freshman grade point average (FGPA from high school GPA and both college admissions and high school tests in mathematics and English. In both systems, the choice of tests had only trivial effects on the aggregate prediction of FGPA. Adding either test to an equation that included the other had only trivial effects on prediction. Although the findings suggest that the choice of test might advantage or disadvantage different students, it had no substantial effect on the over- and underprediction of FGPA for students classified by race-ethnicity or poverty.

  18. Can quantitative sensory testing predict responses to analgesic treatment?

    Science.gov (United States)

    Grosen, K; Fischer, I W D; Olesen, A E; Drewes, A M

    2013-10-01

    The role of quantitative sensory testing (QST) in prediction of analgesic effect in humans is scarcely investigated. This updated review assesses the effectiveness in predicting analgesic effects in healthy volunteers, surgical patients and patients with chronic pain. A systematic review of English written, peer-reviewed articles was conducted using PubMed and Embase (1980-2013). Additional studies were identified by chain searching. Search terms included 'quantitative sensory testing', 'sensory testing' and 'analgesics'. Studies on the relationship between QST and response to analgesic treatment in human adults were included. Appraisal of the methodological quality of the included studies was based on evaluative criteria for prognostic studies. Fourteen studies (including 720 individuals) met the inclusion criteria. Significant correlations were observed between responses to analgesics and several QST parameters including (1) heat pain threshold in experimental human pain, (2) electrical and heat pain thresholds, pressure pain tolerance and suprathreshold heat pain in surgical patients, and (3) electrical and heat pain threshold and conditioned pain modulation in patients with chronic pain. Heterogeneity among studies was observed especially with regard to application of QST and type and use of analgesics. Although promising, the current evidence is not sufficiently robust to recommend the use of any specific QST parameter in predicting analgesic response. Future studies should focus on a range of different experimental pain modalities rather than a single static pain stimulation paradigm. © 2013 European Federation of International Association for the Study of Pain Chapters.

  19. The Prediction of Yarn Elongation of Kenyan Ring-Spun Yarn using Extreme Learning Machines (ELM

    Directory of Open Access Journals (Sweden)

    Josphat Igadwa Mwasiagi

    2017-03-01

    Full Text Available The optimization of the manufacture of cotton yarns involves several processes, while the prediction of yarn quality parameters forms an important area of investigation. This research work concentrated on the prediction of cotton yarn elongation. Cotton lint and yarn samples were collected in textile factories in Kenya.The collected samples were tested under standard testing conditions. Cotton lint parameters, machine parameters and yarn elongation were used to design yarn elongation prediction models. The elongation prediction models used three network training algorithms, including backpropagation (BP, an extreme learning machine (ELM, and a hybrid of differential evolution (DE and an ELM referred to as DE-ELM. The prediction models recorded a mean squared error (mse value of 0.001 using 11, 43 and 2 neurons in the hidden layer for the BP, ELM and DE-ELM models respectively. The ELM models exhibited faster training speeds than the BP algorithms, but required more neurons in the hidden layer than other models. The DEELM hybrid algorithm was faster than the BP algorithm, but slower than the ELM algorithm.

  20. Perceptual Learning Style and Learning Proficiency: A Test of the Hypothesis

    Science.gov (United States)

    Kratzig, Gregory P.; Arbuthnott, Katherine D.

    2006-01-01

    Given the potential importance of using modality preference with instruction, the authors tested whether learning style preference correlated with memory performance in each of 3 sensory modalities: visual, auditory, and kinesthetic. In Study 1, participants completed objective measures of pictorial, auditory, and tactile learning and learning…

  1. The Effect of Formative Testing and Self-Directed Learning on Mathematics Learning Outcomes

    Science.gov (United States)

    Sumantri, Mohamad Syarif; Satriani, Retni

    2016-01-01

    The purpose of this research was to determine the effect of formative testing and self-directed learning on mathematics learning outcomes. The research was conducted at an elementary school in central Jakarta during the 2014/2015 school year. Seventy-two fourth-grade students who were selected using random sampling participated in this study. Data…

  2. Measuring organizational learning. Model testing in two Romanian universities

    OpenAIRE

    Alexandra Luciana Guţă

    2014-01-01

    The scientific literature associates organizational learning with superior organization performance. If we refer to the academic environment, we appreciate that it can develop and reach better levels of performance through changes driven from the inside. Thus, through this paper we elaborate on a conceptual model of organizational learning and we test the model on a sample of employees (university teachers and researchers) from two Romanian universities. The model comprises the process of org...

  3. The ICAP Active Learning Framework Predicts the Learning Gains Observed in Intensely Active Classroom Experiences

    Directory of Open Access Journals (Sweden)

    Benjamin L. Wiggins

    2017-05-01

    Full Text Available STEM classrooms (science, technology, engineering, and mathematics in postsecondary education are rapidly improved by the proper use of active learning techniques. These techniques occupy a descriptive spectrum that transcends passive teaching toward active, constructive, and, finally, interactive methods. While aspects of this framework have been examined, no large-scale or actual classroom-based data exist to inform postsecondary education STEM instructors about possible learning gains. We describe the results of a quasi-experimental study to test the apex of the ICAP framework (interactive, constructive, active, and passive in this ecological classroom environment. Students in interactive classrooms demonstrate significantly improved learning outcomes relative to students in constructive classrooms. This improvement in learning is relatively subtle; similar experimental designs without repeated measures would be unlikely to have the power to observe this significance. We discuss the importance of seemingly small learning gains that might propagate throughout a course or departmental curriculum, as well as improvements with the necessity for faculty to develop and implement similar activities.

  4. Computational prediction of multidisciplinary team decision-making for adjuvant breast cancer drug therapies: a machine learning approach.

    Science.gov (United States)

    Lin, Frank P Y; Pokorny, Adrian; Teng, Christina; Dear, Rachel; Epstein, Richard J

    2016-12-01

    Multidisciplinary team (MDT) meetings are used to optimise expert decision-making about treatment options, but such expertise is not digitally transferable between centres. To help standardise medical decision-making, we developed a machine learning model designed to predict MDT decisions about adjuvant breast cancer treatments. We analysed MDT decisions regarding adjuvant systemic therapy for 1065 breast cancer cases over eight years. Machine learning classifiers with and without bootstrap aggregation were correlated with MDT decisions (recommended, not recommended, or discussable) regarding adjuvant cytotoxic, endocrine and biologic/targeted therapies, then tested for predictability using stratified ten-fold cross-validations. The predictions so derived were duly compared with those based on published (ESMO and NCCN) cancer guidelines. Machine learning more accurately predicted adjuvant chemotherapy MDT decisions than did simple application of guidelines. No differences were found between MDT- vs. ESMO/NCCN- based decisions to prescribe either adjuvant endocrine (97%, p = 0.44/0.74) or biologic/targeted therapies (98%, p = 0.82/0.59). In contrast, significant discrepancies were evident between MDT- and guideline-based decisions to prescribe chemotherapy (87%, p machine learning models. A machine learning approach based on clinicopathologic characteristics can predict MDT decisions about adjuvant breast cancer drug therapies. The discrepancy between MDT- and guideline-based decisions regarding adjuvant chemotherapy implies that certain non-clincopathologic criteria, such as patient preference and resource availability, are factored into clinical decision-making by local experts but not captured by guidelines.

  5. Secondary Structure Prediction of Protein using Resilient Back Propagation Learning Algorithm

    Directory of Open Access Journals (Sweden)

    Jyotshna Dongardive

    2015-12-01

    Full Text Available The paper proposes a neural network based approach to predict secondary structure of protein. It uses Multilayer Feed Forward Network (MLFN with resilient back propagation as the learning algorithm. Point Accepted Mutation (PAM is adopted as the encoding scheme and CB396 data set is used for the training and testing of the network. Overall accuracy of the network has been experimentally calculated with different window sizes for the sliding window scheme and by varying the number of units in the hidden layer. The best results were obtained with eleven as the window size and seven as the number of units in the hidden layer.

  6. Memory Binding Test Predicts Incident Amnestic Mild Cognitive Impairment.

    Science.gov (United States)

    Mowrey, Wenzhu B; Lipton, Richard B; Katz, Mindy J; Ramratan, Wendy S; Loewenstein, David A; Zimmerman, Molly E; Buschke, Herman

    2016-07-14

    The Memory Binding Test (MBT), previously known as Memory Capacity Test, has demonstrated discriminative validity for distinguishing persons with amnestic mild cognitive impairment (aMCI) and dementia from cognitively normal elderly. We aimed to assess the predictive validity of the MBT for incident aMCI. In a longitudinal, community-based study of adults aged 70+, we administered the MBT to 246 cognitively normal elderly adults at baseline and followed them annually. Based on previous work, a subtle reduction in memory binding at baseline was defined by a Total Items in the Paired (TIP) condition score of ≤22 on the MBT. Cox proportional hazards models were used to assess the predictive validity of the MBT for incident aMCI accounting for the effects of covariates. The hazard ratio of incident aMCI was also assessed for different prediction time windows ranging from 4 to 7 years of follow-up, separately. Among 246 controls who were cognitively normal at baseline, 48 developed incident aMCI during follow-up. A baseline MBT reduction was associated with an increased risk for developing incident aMCI (hazard ratio (HR) = 2.44, 95% confidence interval: 1.30-4.56, p = 0.005). When varying the prediction window from 4-7 years, the MBT reduction remained significant for predicting incident aMCI (HR range: 2.33-3.12, p: 0.0007-0.04). Persons with poor performance on the MBT are at significantly greater risk for developing incident aMCI. High hazard ratios up to seven years of follow-up suggest that the MBT is sensitive to early disease.

  7. Controlled test for predictive power of Lyapunov exponents: their inability to predict epileptic seizures.

    Science.gov (United States)

    Lai, Ying-Cheng; Harrison, Mary Ann F; Frei, Mark G; Osorio, Ivan

    2004-09-01

    Lyapunov exponents are a set of fundamental dynamical invariants characterizing a system's sensitive dependence on initial conditions. For more than a decade, it has been claimed that the exponents computed from electroencephalogram (EEG) or electrocorticogram (ECoG) signals can be used for prediction of epileptic seizures minutes or even tens of minutes in advance. The purpose of this paper is to examine the predictive power of Lyapunov exponents. Three approaches are employed. (1) We present qualitative arguments suggesting that the Lyapunov exponents generally are not useful for seizure prediction. (2) We construct a two-dimensional, nonstationary chaotic map with a parameter slowly varying in a range containing a crisis, and test whether this critical event can be predicted by monitoring the evolution of finite-time Lyapunov exponents. This can thus be regarded as a "control test" for the claimed predictive power of the exponents for seizure. We find that two major obstacles arise in this application: statistical fluctuations of the Lyapunov exponents due to finite time computation and noise from the time series. We show that increasing the amount of data in a moving window will not improve the exponents' detective power for characteristic system changes, and that the presence of small noise can ruin completely the predictive power of the exponents. (3) We report negative results obtained from ECoG signals recorded from patients with epilepsy. All these indicate firmly that, the use of Lyapunov exponents for seizure prediction is practically impossible as the brain dynamical system generating the ECoG signals is more complicated than low-dimensional chaotic systems, and is noisy. Copyright 2004 American Institute of Physics

  8. The Prediction of Students' Academic Performance With Fluid Intelligence in Giving Special Consideration to the Contribution of Learning.

    Science.gov (United States)

    Ren, Xuezhu; Schweizer, Karl; Wang, Tengfei; Xu, Fen

    2015-01-01

    The present study provides a new account of how fluid intelligence influences academic performance. In this account a complex learning component of fluid intelligence tests is proposed to play a major role in predicting academic performance. A sample of 2, 277 secondary school students completed two reasoning tests that were assumed to represent fluid intelligence and standardized math and verbal tests assessing academic performance. The fluid intelligence data were decomposed into a learning component that was associated with the position effect of intelligence items and a constant component that was independent of the position effect. Results showed that the learning component contributed significantly more to the prediction of math and verbal performance than the constant component. The link from the learning component to math performance was especially strong. These results indicated that fluid intelligence, which has so far been considered as homogeneous, could be decomposed in such a way that the resulting components showed different properties and contributed differently to the prediction of academic performance. Furthermore, the results were in line with the expectation that learning was a predictor of performance in school.

  9. Learning Combinations of Multiple Feature Representations for Music Emotion Prediction

    DEFF Research Database (Denmark)

    Madsen, Jens; Jensen, Bjørn Sand; Larsen, Jan

    2015-01-01

    Music consists of several structures and patterns evolving through time which greatly influences the human decoding of higher-level cognitive aspects of music like the emotions expressed in music. For tasks, such as genre, tag and emotion recognition, these structures have often been identified...... and used as individual and non-temporal features and representations. In this work, we address the hypothesis whether using multiple temporal and non-temporal representations of different features is beneficial for modeling music structure with the aim to predict the emotions expressed in music. We test...

  10. Developing robust arsenic awareness prediction models using machine learning algorithms.

    Science.gov (United States)

    Singh, Sushant K; Taylor, Robert W; Rahman, Mohammad Mahmudur; Pradhan, Biswajeet

    2018-04-01

    Arsenic awareness plays a vital role in ensuring the sustainability of arsenic mitigation technologies. Thus far, however, few studies have dealt with the sustainability of such technologies and its associated socioeconomic dimensions. As a result, arsenic awareness prediction has not yet been fully conceptualized. Accordingly, this study evaluated arsenic awareness among arsenic-affected communities in rural India, using a structured questionnaire to record socioeconomic, demographic, and other sociobehavioral factors with an eye to assessing their association with and influence on arsenic awareness. First a logistic regression model was applied and its results compared with those produced by six state-of-the-art machine-learning algorithms (Support Vector Machine [SVM], Kernel-SVM, Decision Tree [DT], k-Nearest Neighbor [k-NN], Naïve Bayes [NB], and Random Forests [RF]) as measured by their accuracy at predicting arsenic awareness. Most (63%) of the surveyed population was found to be arsenic-aware. Significant arsenic awareness predictors were divided into three types: (1) socioeconomic factors: caste, education level, and occupation; (2) water and sanitation behavior factors: number of family members involved in water collection, distance traveled and time spent for water collection, places for defecation, and materials used for handwashing after defecation; and (3) social capital and trust factors: presence of anganwadi and people's trust in other community members, NGOs, and private agencies. Moreover, individuals' having higher social network positively contributed to arsenic awareness in the communities. Results indicated that both the SVM and the RF algorithms outperformed at overall prediction of arsenic awareness-a nonlinear classification problem. Lower-caste, less educated, and unemployed members of the population were found to be the most vulnerable, requiring immediate arsenic mitigation. To this end, local social institutions and NGOs could play a

  11. Effectiveness of Cooperative Learning Instructional Tools With Predict-Observe-Explain Strategy on the Topic of Cuboid and Cube Volume

    Science.gov (United States)

    Nurhuda; Lukito, A.; Masriyah

    2018-01-01

    This study aims to develop instructional tools and implement it to see the effectiveness. The method used in this research referred to Designing Effective Instruction. Experimental research with two-group pretest-posttest design method was conducted. The instructional tools have been developed is cooperative learning model with predict-observe-explain strategy on the topic of cuboid and cube volume which consist of lesson plans, POE tasks, and Tests. Instructional tools were of good quality by criteria of validity, practicality, and effectiveness. These instructional tools was very effective for teaching the volume of cuboid and cube. Cooperative instructional tool with predict-observe-explain (POE) strategy was good of quality because the teacher was easy to implement the steps of learning, students easy to understand the material and students’ learning outcomes completed classically. Learning by using this instructional tool was effective because learning activities were appropriate and students were very active. Students’ learning outcomes were completed classically and better than conventional learning. This study produced a good instructional tool and effectively used in learning. Therefore, these instructional tools can be used as an alternative to teach volume of cuboid and cube topics.

  12. Modeling learning and memory using verbal learning tests: results from ACTIVE.

    Science.gov (United States)

    Gross, Alden L; Rebok, George W; Brandt, Jason; Tommet, Doug; Marsiske, Michael; Jones, Richard N

    2013-03-01

    To investigate the influence of memory training on initial recall and learning. The Advanced Cognitive Training for Independent and Vital Elderly study of community-dwelling adults older than age 65 (n = 1,401). We decomposed trial-level recall in the Auditory Verbal Learning Test (AVLT) and Hopkins Verbal Learning Test (HVLT) into initial recall and learning across trials using latent growth models. Trial-level increases in words recalled in the AVLT and HVLT at each follow-up visit followed an approximately logarithmic shape. Over the 5-year study period, memory training was associated with slower decline in Trial 1 AVLT recall (Cohen's d = 0.35, p = .03) and steep pre- and posttraining acceleration in learning (d = 1.56, p learning, d = 3.10, p memory-trained group had a higher level of recall than the control group through the end of the 5-year study period despite faster decline in learning. This study contributes to the understanding of the mechanisms by which training benefits memory and expands current knowledge by reporting long-term changes in initial recall and learning, as measured from growth models and by characterization of the impact of memory training on these components. Results reveal that memory training delays the worsening of memory span and boosts learning.

  13. Ethical issues in predictive genetic testing: a public health perspective

    Science.gov (United States)

    Fulda, K G; Lykens, K

    2006-01-01

    As a result of the increase in genetic testing and the fear of discrimination by insurance companies, employers, and society as a result of genetic testing, the disciplines of ethics, public health, and genetics have converged. Whether relatives of someone with a positive predictive genetic test should be notified of the results and risks is a matter urgently in need of debate. Such a debate must encompass the moral and ethical obligations of the diagnosing physician and the patient. The decision to inform or not will vary depending on what moral theory is used. Utilising the utilitarian and libertarian theories produces different outcomes. The principles of justice and non‐maleficence will also play an important role in the decision. PMID:16507657

  14. Ethical issues in predictive genetic testing: a public health perspective.

    Science.gov (United States)

    Fulda, K G; Lykens, K

    2006-03-01

    As a result of the increase in genetic testing and the fear of discrimination by insurance companies, employers, and society as a result of genetic testing, the disciplines of ethics, public health, and genetics have converged. Whether relatives of someone with a positive predictive genetic test should be notified of the results and risks is a matter urgently in need of debate. Such a debate must encompass the moral and ethical obligations of the diagnosing physician and the patient. The decision to inform or not will vary depending on what moral theory is used. Utilising the utilitarian and libertarian theories produces different outcomes. The principles of justice and non-maleficence will also play an important role in the decision.

  15. Observing others stay or switch - How social prediction errors are integrated into reward reversal learning.

    Science.gov (United States)

    Ihssen, Niklas; Mussweiler, Thomas; Linden, David E J

    2016-08-01

    Reward properties of stimuli can undergo sudden changes, and the detection of these 'reversals' is often made difficult by the probabilistic nature of rewards/punishments. Here we tested whether and how humans use social information (someone else's choices) to overcome uncertainty during reversal learning. We show a substantial social influence during reversal learning, which was modulated by the type of observed behavior. Participants frequently followed observed conservative choices (no switches after punishment) made by the (fictitious) other player but ignored impulsive choices (switches), even though the experiment was set up so that both types of response behavior would be similarly beneficial/detrimental (Study 1). Computational modeling showed that participants integrated the observed choices as a 'social prediction error' instead of ignoring or blindly following the other player. Modeling also confirmed higher learning rates for 'conservative' versus 'impulsive' social prediction errors. Importantly, this 'conservative bias' was boosted by interpersonal similarity, which in conjunction with the lack of effects observed in a non-social control experiment (Study 2) confirmed its social nature. A third study suggested that relative weighting of observed impulsive responses increased with increased volatility (frequency of reversals). Finally, simulations showed that in the present paradigm integrating social and reward information was not necessarily more adaptive to maximize earnings than learning from reward alone. Moreover, integrating social information increased accuracy only when conservative and impulsive choices were weighted similarly during learning. These findings suggest that to guide decisions in choice contexts that involve reward reversals humans utilize social cues conforming with their preconceptions more strongly than cues conflicting with them, especially when the other is similar. Copyright © 2016 The Authors. Published by Elsevier B

  16. Days on radiosensitivity: individual variability and predictive tests; Radiosensibilite: variabilite individuelle et tests predictifs

    Energy Technology Data Exchange (ETDEWEB)

    NONE

    2008-07-01

    The radiosensitivity is a part of usual clinical observations. It is already included in the therapy protocols. however, some questions stay on its individual variability and on the difficulty to evaluate it. The point will be stocked on its origin and its usefulness in predictive medicine. Through examples on the use of predictive tests and ethical and legal questions that they raise, concrete cases will be presented by specialists such radio biologists, geneticists, immunologists, jurists and occupational physicians. (N.C.)

  17. Predictive Coding Accelerates Word Recognition and Learning in the Early Stages of Language Development

    Science.gov (United States)

    Ylinen, Sari; Bosseler, Alexis; Junttila, Katja; Huotilainen, Minna

    2017-01-01

    The ability to predict future events in the environment and learn from them is a fundamental component of adaptive behavior across species. Here we propose that inferring predictions facilitates speech processing and word learning in the early stages of language development. Twelve- and 24-month olds' electrophysiological brain responses to heard…

  18. Scaling prediction errors to reward variability benefits error-driven learning in humans.

    Science.gov (United States)

    Diederen, Kelly M J; Schultz, Wolfram

    2015-09-01

    Effective error-driven learning requires individuals to adapt learning to environmental reward variability. The adaptive mechanism may involve decays in learning rate across subsequent trials, as shown previously, and rescaling of reward prediction errors. The present study investigated the influence of prediction error scaling and, in particular, the consequences for learning performance. Participants explicitly predicted reward magnitudes that were drawn from different probability distributions with specific standard deviations. By fitting the data with reinforcement learning models, we found scaling of prediction errors, in addition to the learning rate decay shown previously. Importantly, the prediction error scaling was closely related to learning performance, defined as accuracy in predicting the mean of reward distributions, across individual participants. In addition, participants who scaled prediction errors relative to standard deviation also presented with more similar performance for different standard deviations, indicating that increases in standard deviation did not substantially decrease "adapters'" accuracy in predicting the means of reward distributions. However, exaggerated scaling beyond the standard deviation resulted in impaired performance. Thus efficient adaptation makes learning more robust to changing variability. Copyright © 2015 the American Physiological Society.

  19. LEARNING AND MEMORY TESTS IN DEVELOPMENTAL NEUROTOXICITY TESTING: A CROSS-LABORATORY COMPARISON OF CONTROL DATA.

    Science.gov (United States)

    The US EPA Developmental Neurotoxicity (DNT) Study Test Guideline (OPPTS 870.6300) calls for functional tests to assess the impact of chemicals on cognitive function in offspring following maternal exposure. A test of associative learning and memory is to be conducted around th...

  20. A "Uses and Gratification Expectancy Model" to Predict Students' "Perceived e-Learning Experience"

    Science.gov (United States)

    Mondi, Makingu; Woods, Peter; Rafi, Ahmad

    2008-01-01

    This study investigates "how and why" students' "Uses and Gratification Expectancy" (UGE) for e-learning resources influences their "Perceived e-Learning Experience." A "Uses and Gratification Expectancy Model" (UGEM) framework is proposed to predict students' "Perceived e-Learning Experience," and…

  1. Learning effect and test-retest variability of pulsar perimetry.

    Science.gov (United States)

    Salvetat, Maria Letizia; Zeppieri, Marco; Parisi, Lucia; Johnson, Chris A; Sampaolesi, Roberto; Brusini, Paolo

    2013-03-01

    To assess Pulsar Perimetry learning effect and test-retest variability (TRV) in normal (NORM), ocular hypertension (OHT), glaucomatous optic neuropathy (GON), and primary open-angle glaucoma (POAG) eyes. This multicenter prospective study included 43 NORM, 38 OHT, 33 GON, and 36 POAG patients. All patients underwent standard automated perimetry and Pulsar Contrast Perimetry using white stimuli modulated in phase and counterphase at 30 Hz (CP-T30W test). The learning effect and TRV for Pulsar Perimetry were assessed for 3 consecutive visual fields (VFs). The learning effect were evaluated by comparing results from the first session with the other 2. TRV was assessed by calculating the mean of the differences (in absolute value) between retests for each combination of single tests. TRV was calculated for Mean Sensitivity, Mean Defect, and single Mean Sensitivity for each 66 test locations. Influence of age, VF eccentricity, and loss severity on TRV were assessed using linear regression analysis and analysis of variance. The learning effect was not significant in any group (analysis of variance, P>0.05). TRV for Mean Sensitivity and Mean Defect was significantly lower in NORM and OHT (0.6 ± 0.5 spatial resolution contrast units) than in GON and POAG (0.9 ± 0.5 and 1.0 ± 0.8 spatial resolution contrast units, respectively) (Kruskal-Wallis test, P=0.04); however, the differences in NORM among age groups was not significant (Kruskal-Wallis test, P>0.05). Slight significant differences were found for the single Mean Sensitivity TRV among single locations (Duncan test, PPulsar Perimetry CP-T30W test did not show significant learning effect in patients with standard automated perimetry experience. TRV for global indices was generally low, and was not related to patient age; it was only slightly affected by VF defect eccentricity, and significantly influenced by VF loss severity.

  2. Machine learning applications in cancer prognosis and prediction.

    Science.gov (United States)

    Kourou, Konstantina; Exarchos, Themis P; Exarchos, Konstantinos P; Karamouzis, Michalis V; Fotiadis, Dimitrios I

    2015-01-01

    Cancer has been characterized as a heterogeneous disease consisting of many different subtypes. The early diagnosis and prognosis of a cancer type have become a necessity in cancer research, as it can facilitate the subsequent clinical management of patients. The importance of classifying cancer patients into high or low risk groups has led many research teams, from the biomedical and the bioinformatics field, to study the application of machine learning (ML) methods. Therefore, these techniques have been utilized as an aim to model the progression and treatment of cancerous conditions. In addition, the ability of ML tools to detect key features from complex datasets reveals their importance. A variety of these techniques, including Artificial Neural Networks (ANNs), Bayesian Networks (BNs), Support Vector Machines (SVMs) and Decision Trees (DTs) have been widely applied in cancer research for the development of predictive models, resulting in effective and accurate decision making. Even though it is evident that the use of ML methods can improve our understanding of cancer progression, an appropriate level of validation is needed in order for these methods to be considered in the everyday clinical practice. In this work, we present a review of recent ML approaches employed in the modeling of cancer progression. The predictive models discussed here are based on various supervised ML techniques as well as on different input features and data samples. Given the growing trend on the application of ML methods in cancer research, we present here the most recent publications that employ these techniques as an aim to model cancer risk or patient outcomes.

  3. HPV-testing versus HPV-cytology co-testing to predict the outcome after conization.

    Science.gov (United States)

    Bruhn, Laerke Valsøe; Andersen, Sisse Josephine; Hariri, Jalil

    2018-06-01

    The purpose of this study was to determine the feasibility of human Papillomavirus (HPV) testing alone as a prognostic tool to predict recurrent disease within a three-year follow-up period after treatment for cervical intraepithelial neoplasia (CIN)2 + . Retrospectively, 128 women with histologically verified CIN2 + who had a conization performed at Southern Jutland Hospital in Denmark between 1 January 2013 and 31 December 2013 were included. Histology, cytology and HPV test results were obtained for a three-year follow-up period. 4.7% (6/128) of the cases developed recurrent disease during follow-up. Of the cases without free margins, recurrent dysplasia was detected normal in 10.4% (5/48), whereas in the group with free margins it was 1.3% (1/80). The post-conization HPV test was negative in 67.2% (86/128) and Pap smear normal in 93.7% (120/128). Combining resection margins, cytology and HPV had sensitivity for prediction of recurrent dysplasia of 100%. Specificity was 45.8%, positive predictive value (PPV) 8.5% and negative predictive value (NPV) 100%. Using HPV test alone as a predictor of recurrent dysplasia gave a sensitivity of 83.3%, specificity 69.7%, PPV 11.9% and NPV 98.8%. Combining resection margin and HPV test had a sensitivity of 100%, specificity 45.9%, PPV 8.3% and NPV 100%. HPV test at six months control post-conization gave an NPV of 98.8% and can be used as a solitary test to identify women at risk for recurrent disease three years after treatment for precursor lesions. Using both resection margin and HPV test had a sensitivity of 100% and NPV 100%. Adding cytology did not increase the predictive value. © 2018 Nordic Federation of Societies of Obstetrics and Gynecology.

  4. Surprised at all the entropy: hippocampal, caudate and midbrain contributions to learning from prediction errors.

    Directory of Open Access Journals (Sweden)

    Anne-Marike Schiffer

    Full Text Available Influential concepts in neuroscientific research cast the brain a predictive machine that revises its predictions when they are violated by sensory input. This relates to the predictive coding account of perception, but also to learning. Learning from prediction errors has been suggested for take place in the hippocampal memory system as well as in the basal ganglia. The present fMRI study used an action-observation paradigm to investigate the contributions of the hippocampus, caudate nucleus and midbrain dopaminergic system to different types of learning: learning in the absence of prediction errors, learning from prediction errors, and responding to the accumulation of prediction errors in unpredictable stimulus configurations. We conducted analyses of the regions of interests' BOLD response towards these different types of learning, implementing a bootstrapping procedure to correct for false positives. We found both, caudate nucleus and the hippocampus to be activated by perceptual prediction errors. The hippocampal responses seemed to relate to the associative mismatch between a stored representation and current sensory input. Moreover, its response was significantly influenced by the average information, or Shannon entropy of the stimulus material. In accordance with earlier results, the habenula was activated by perceptual prediction errors. Lastly, we found that the substantia nigra was activated by the novelty of sensory input. In sum, we established that the midbrain dopaminergic system, the hippocampus, and the caudate nucleus were to different degrees significantly involved in the three different types of learning: acquisition of new information, learning from prediction errors and responding to unpredictable stimulus developments. We relate learning from perceptual prediction errors to the concept of predictive coding and related information theoretic accounts.

  5. Surprised at all the entropy: hippocampal, caudate and midbrain contributions to learning from prediction errors.

    Science.gov (United States)

    Schiffer, Anne-Marike; Ahlheim, Christiane; Wurm, Moritz F; Schubotz, Ricarda I

    2012-01-01

    Influential concepts in neuroscientific research cast the brain a predictive machine that revises its predictions when they are violated by sensory input. This relates to the predictive coding account of perception, but also to learning. Learning from prediction errors has been suggested for take place in the hippocampal memory system as well as in the basal ganglia. The present fMRI study used an action-observation paradigm to investigate the contributions of the hippocampus, caudate nucleus and midbrain dopaminergic system to different types of learning: learning in the absence of prediction errors, learning from prediction errors, and responding to the accumulation of prediction errors in unpredictable stimulus configurations. We conducted analyses of the regions of interests' BOLD response towards these different types of learning, implementing a bootstrapping procedure to correct for false positives. We found both, caudate nucleus and the hippocampus to be activated by perceptual prediction errors. The hippocampal responses seemed to relate to the associative mismatch between a stored representation and current sensory input. Moreover, its response was significantly influenced by the average information, or Shannon entropy of the stimulus material. In accordance with earlier results, the habenula was activated by perceptual prediction errors. Lastly, we found that the substantia nigra was activated by the novelty of sensory input. In sum, we established that the midbrain dopaminergic system, the hippocampus, and the caudate nucleus were to different degrees significantly involved in the three different types of learning: acquisition of new information, learning from prediction errors and responding to unpredictable stimulus developments. We relate learning from perceptual prediction errors to the concept of predictive coding and related information theoretic accounts.

  6. Prediction of software operational reliability using testing environment factor

    International Nuclear Information System (INIS)

    Jung, Hoan Sung

    1995-02-01

    Software reliability is especially important to customers these days. The need to quantify software reliability of safety-critical systems has been received very special attention and the reliability is rated as one of software's most important attributes. Since the software is an intellectual product of human activity and since it is logically complex, the failures are inevitable. No standard models have been established to prove the correctness and to estimate the reliability of software systems by analysis and/or testing. For many years, many researches have focused on the quantification of software reliability and there are many models developed to quantify software reliability. Most software reliability models estimate the reliability with the failure data collected during the test assuming that the test environments well represent the operation profile. User's interest is on the operational reliability rather than on the test reliability, however. The experiences show that the operational reliability is higher than the test reliability. With the assumption that the difference in reliability results from the change of environment, testing environment factor comprising the aging factor and the coverage factor are defined in this work to predict the ultimate operational reliability with the failure data. It is by incorporating test environments applied beyond the operational profile into testing environment factor Test reliability can also be estimated with this approach without any model change. The application results are close to the actual data. The approach used in this thesis is expected to be applicable to ultra high reliable software systems that are used in nuclear power plants, airplanes, and other safety-critical applications

  7. Supervised learning classification models for prediction of plant virus encoded RNA silencing suppressors.

    Directory of Open Access Journals (Sweden)

    Zeenia Jagga

    Full Text Available Viral encoded RNA silencing suppressor proteins interfere with the host RNA silencing machinery, facilitating viral infection by evading host immunity. In plant hosts, the viral proteins have several basic science implications and biotechnology applications. However in silico identification of these proteins is limited by their high sequence diversity. In this study we developed supervised learning based classification models for plant viral RNA silencing suppressor proteins in plant viruses. We developed four classifiers based on supervised learning algorithms: J48, Random Forest, LibSVM and Naïve Bayes algorithms, with enriched model learning by correlation based feature selection. Structural and physicochemical features calculated for experimentally verified primary protein sequences were used to train the classifiers. The training features include amino acid composition; auto correlation coefficients; composition, transition, and distribution of various physicochemical properties; and pseudo amino acid composition. Performance analysis of predictive models based on 10 fold cross-validation and independent data testing revealed that the Random Forest based model was the best and achieved 86.11% overall accuracy and 86.22% balanced accuracy with a remarkably high area under the Receivers Operating Characteristic curve of 0.95 to predict viral RNA silencing suppressor proteins. The prediction models for plant viral RNA silencing suppressors can potentially aid identification of novel viral RNA silencing suppressors, which will provide valuable insights into the mechanism of RNA silencing and could be further explored as potential targets for designing novel antiviral therapeutics. Also, the key subset of identified optimal features may help in determining compositional patterns in the viral proteins which are important determinants for RNA silencing suppressor activities. The best prediction model developed in the study is available as a

  8. Predicting Pre-planting Risk of Stagonospora nodorum blotch in Winter Wheat Using Machine Learning Models

    Directory of Open Access Journals (Sweden)

    Lucky eMehra

    2016-03-01

    Full Text Available Pre-planting factors have been associated with the late-season severity of Stagonospora nodorum blotch (SNB, caused by the fungal pathogen Parastagonospora nodorum, in winter wheat (Triticum aestivum. The relative importance of these factors in the risk of SNB has not been determined and this knowledge can facilitate disease management decisions prior to planting of the wheat crop. In this study, we examined the performance of multiple regression (MR and three machine learning algorithms namely artificial neural networks, categorical and regression trees, and random forests (RF in predicting the pre-planting risk of SNB in wheat. Pre-planting factors tested as potential predictor variables were cultivar resistance, latitude, longitude, previous crop, seeding rate, seed treatment, tillage type, and wheat residue. Disease severity assessed at the end of the growing season was used as the response variable. The models were developed using 431 disease cases (unique combinations of predictors collected from 2012 to 2014 and these cases were randomly divided into training, validation, and test datasets. Models were evaluated based on the regression of observed against predicted severity values of SNB, sensitivity-specificity ROC analysis, and the Kappa statistic. A strong relationship was observed between late-season severity of SNB and specific pre-planting factors in which latitude, longitude, wheat residue, and cultivar resistance were the most important predictors. The MR model explained 33% of variability in the data, while machine learning models explained 47 to 79% of the total variability. Similarly, the MR model correctly classified 74% of the disease cases, while machine learning models correctly classified 81 to 83% of these cases. Results show that the RF algorithm, which explained 79% of the variability within the data, was the most accurate in predicting the risk of SNB, with an accuracy rate of 93%. The RF algorithm could allow early

  9. Predicting Pre-planting Risk of Stagonospora nodorum blotch in Winter Wheat Using Machine Learning Models.

    Science.gov (United States)

    Mehra, Lucky K; Cowger, Christina; Gross, Kevin; Ojiambo, Peter S

    2016-01-01

    Pre-planting factors have been associated with the late-season severity of Stagonospora nodorum blotch (SNB), caused by the fungal pathogen Parastagonospora nodorum, in winter wheat (Triticum aestivum). The relative importance of these factors in the risk of SNB has not been determined and this knowledge can facilitate disease management decisions prior to planting of the wheat crop. In this study, we examined the performance of multiple regression (MR) and three machine learning algorithms namely artificial neural networks, categorical and regression trees, and random forests (RF), in predicting the pre-planting risk of SNB in wheat. Pre-planting factors tested as potential predictor variables were cultivar resistance, latitude, longitude, previous crop, seeding rate, seed treatment, tillage type, and wheat residue. Disease severity assessed at the end of the growing season was used as the response variable. The models were developed using 431 disease cases (unique combinations of predictors) collected from 2012 to 2014 and these cases were randomly divided into training, validation, and test datasets. Models were evaluated based on the regression of observed against predicted severity values of SNB, sensitivity-specificity ROC analysis, and the Kappa statistic. A strong relationship was observed between late-season severity of SNB and specific pre-planting factors in which latitude, longitude, wheat residue, and cultivar resistance were the most important predictors. The MR model explained 33% of variability in the data, while machine learning models explained 47 to 79% of the total variability. Similarly, the MR model correctly classified 74% of the disease cases, while machine learning models correctly classified 81 to 83% of these cases. Results show that the RF algorithm, which explained 79% of the variability within the data, was the most accurate in predicting the risk of SNB, with an accuracy rate of 93%. The RF algorithm could allow early assessment of

  10. To transfer or not to transfer? Kinematics and laterality quotient predict interlimb transfer of motor learning.

    Science.gov (United States)

    Lefumat, Hannah Z; Vercher, Jean-Louis; Miall, R Chris; Cole, Jonathan; Buloup, Frank; Bringoux, Lionel; Bourdin, Christophe; Sarlegna, Fabrice R

    2015-11-01

    Humans can remarkably adapt their motor behavior to novel environmental conditions, yet it remains unclear which factors enable us to transfer what we have learned with one limb to the other. Here we tested the hypothesis that interlimb transfer of sensorimotor adaptation is determined by environmental conditions but also by individual characteristics. We specifically examined the adaptation of unconstrained reaching movements to a novel Coriolis, velocity-dependent force field. Right-handed subjects sat at the center of a rotating platform and performed forward reaching movements with the upper limb toward flashed visual targets in prerotation, per-rotation (i.e., adaptation), and postrotation tests. Here only the dominant arm was used during adaptation and interlimb transfer was assessed by comparing performance of the nondominant arm before and after dominant-arm adaptation. Vision and no-vision conditions did not significantly influence interlimb transfer of trajectory adaptation, which on average was significant but limited. We uncovered a substantial heterogeneity of interlimb transfer across subjects and found that interlimb transfer can be qualitatively and quantitatively predicted for each healthy young individual. A classifier showed that in our study, interlimb transfer could be predicted based on the subject's task performance, most notably motor variability during learning, and his or her laterality quotient. Positive correlations suggested that variability of motor performance and lateralization of arm movement control facilitate interlimb transfer. We further show that these individual characteristics can predict the presence and the magnitude of interlimb transfer of left-handers. Overall, this study suggests that individual characteristics shape the way the nervous system can generalize motor learning. Copyright © 2015 the American Physiological Society.

  11. Predicting novel substrates for enzymes with minimal experimental effort with active learning.

    Science.gov (United States)

    Pertusi, Dante A; Moura, Matthew E; Jeffryes, James G; Prabhu, Siddhant; Walters Biggs, Bradley; Tyo, Keith E J

    2017-11-01

    Enzymatic substrate promiscuity is more ubiquitous than previously thought, with significant consequences for understanding metabolism and its application to biocatalysis. This realization has given rise to the need for efficient characterization of enzyme promiscuity. Enzyme promiscuity is currently characterized with a limited number of human-selected compounds that may not be representative of the enzyme's versatility. While testing large numbers of compounds may be impractical, computational approaches can exploit existing data to determine the most informative substrates to test next, thereby more thoroughly exploring an enzyme's versatility. To demonstrate this, we used existing studies and tested compounds for four different enzymes, developed support vector machine (SVM) models using these datasets, and selected additional compounds for experiments using an active learning approach. SVMs trained on a chemically diverse set of compounds were discovered to achieve maximum accuracies of ~80% using ~33% fewer compounds than datasets based on all compounds tested in existing studies. Active learning-selected compounds for testing resolved apparent conflicts in the existing training data, while adding diversity to the dataset. The application of these algorithms to wide arrays of metabolic enzymes would result in a library of SVMs that can predict high-probability promiscuous enzymatic reactions and could prove a valuable resource for the design of novel metabolic pathways. Copyright © 2017 International Metabolic Engineering Society. Published by Elsevier Inc. All rights reserved.

  12. Predicting novel substrates for enzymes with minimal experimental effort with active learning

    Energy Technology Data Exchange (ETDEWEB)

    Pertusi, Dante A.; Moura, Matthew E.; Jeffryes, James G.; Prabhu, Siddhant; Walters Biggs, Bradley; Tyo, Keith E. J.

    2017-11-01

    Enzymatic substrate promiscuity is more ubiquitous than previously thought, with significant consequences for understanding metabolism and its application to biocatalysis. This realization has given rise to the need for efficient characterization of enzyme promiscuity. Enzyme promiscuity is currently characterized with a limited number of human-selected compounds that may not be representative of the enzyme's versatility. While testing large numbers of compounds may be impractical, computational approaches can exploit existing data to determine the most informative substrates to test next, thereby more thoroughly exploring an enzyme's versatility. To demonstrate this, we used existing studies and tested compounds for four different enzymes, developed support vector machine (SVM) models using these datasets, and selected additional compounds for experiments using an active learning approach. SVMs trained on a chemically diverse set of compounds were discovered to achieve maximum accuracies of similar to 80% using similar to 33% fewer compounds than datasets based on all compounds tested in existing studies. Active learning-selected compounds for testing resolved apparent conflicts in the existing training data, while adding diversity to the dataset. The application of these algorithms to wide arrays of metabolic enzymes would result in a library of SVMs that can predict high-probability promiscuous enzymatic reactions and could prove a valuable resource for the design of novel metabolic pathways.

  13. DeepRT: deep learning for peptide retention time prediction in proteomics

    OpenAIRE

    Ma, Chunwei; Zhu, Zhiyong; Ye, Jun; Yang, Jiarui; Pei, Jianguo; Xu, Shaohang; Zhou, Ruo; Yu, Chang; Mo, Fan; Wen, Bo; Liu, Siqi

    2017-01-01

    Accurate predictions of peptide retention times (RT) in liquid chromatography have many applications in mass spectrometry-based proteomics. Herein, we present DeepRT, a deep learning based software for peptide retention time prediction. DeepRT automatically learns features directly from the peptide sequences using the deep convolutional Neural Network (CNN) and Recurrent Neural Network (RNN) model, which eliminates the need to use hand-crafted features or rules. After the feature learning, pr...

  14. Machine Learning and Deep Learning Models to Predict Runoff Water Quantity and Quality

    Science.gov (United States)

    Bradford, S. A.; Liang, J.; Li, W.; Murata, T.; Simunek, J.

    2017-12-01

    Contaminants can be rapidly transported at the soil surface by runoff to surface water bodies. Physically-based models, which are based on the mathematical description of main hydrological processes, are key tools for predicting surface water impairment. Along with physically-based models, data-driven models are becoming increasingly popular for describing the behavior of hydrological and water resources systems since these models can be used to complement or even replace physically based-models. In this presentation we propose a new data-driven model as an alternative to a physically-based overland flow and transport model. First, we have developed a physically-based numerical model to simulate overland flow and contaminant transport (the HYDRUS-1D overland flow module). A large number of numerical simulations were carried out to develop a database containing information about the impact of various input parameters (weather patterns, surface topography, vegetation, soil conditions, contaminants, and best management practices) on runoff water quantity and quality outputs. This database was used to train data-driven models. Three different methods (Neural Networks, Support Vector Machines, and Recurrence Neural Networks) were explored to prepare input- output functional relations. Results demonstrate the ability and limitations of machine learning and deep learning models to predict runoff water quantity and quality.

  15. AGR-5/6/7 Irradiation Test Predictions using PARFUME

    Energy Technology Data Exchange (ETDEWEB)

    Skerjanc, William F. [Idaho National Lab. (INL), Idaho Falls, ID (United States)

    2017-09-14

    PARFUME, (PARticle FUel ModEl) a fuel performance modeling code used for high temperature gas-cooled reactors (HTGRs), was used to model the Advanced Gas Reactor (AGR)-5/6/7 irradiation test using predicted physics and thermal hydraulics data. The AGR-5/6/7 test consists of the combined fifth, sixth, and seventh planned irradiations of the AGR Fuel Development and Qualification Program. The AGR-5/6/7 test train is a multi-capsule, instrumented experiment that is designed for irradiation in the 133.4-mm diameter north east flux trap (NEFT) position of Advanced Test Reactor (ATR). Each capsule contains compacts filled with uranium oxycarbide (UCO) unaltered fuel particles. This report documents the calculations performed to predict the failure probability of tristructural isotropic (TRISO)-coated fuel particles during the AGR-5/6/7 experiment. In addition, this report documents the calculated source term from the driver fuel. The calculations include modeling of the AGR-5/6/7 irradiation that is scheduled to occur from October 2017 to April 2021 over a total of 13 ATR cycles, including nine normal cycles and four Power Axial Locator Mechanism (PALM) cycle for a total between 500 – 550 effective full power days (EFPD). The irradiation conditions and material properties of the AGR-5/6/7 test predicted zero fuel particle failures in Capsules 1, 2, and 4. Fuel particle failures were predicted in Capsule 3 due to internal particle pressure. These failures were predicted in the highest temperature compacts. Capsule 5 fuel particle failures were due to inner pyrolytic carbon (IPyC) cracking causing localized stresses concentrations in the SiC layer. This capsule predicted the highest particle failures due to the lower irradiation temperature. In addition, shrinkage of the buffer and IPyC layer during irradiation resulted in formation of a buffer-IPyC gap. The two capsules at the two ends of the test train, Capsules 1 and 5 experienced the smallest buffer-IPyC gap

  16. Opportunity to learn: Investigating possible predictors for pre-course Test Of Astronomy STandards TOAST scores

    Science.gov (United States)

    Berryhill, Katie J.

    As astronomy education researchers become more interested in experimentally testing innovative teaching strategies to enhance learning in introductory astronomy survey courses ("ASTRO 101"), scholars are placing increased attention toward better understanding factors impacting student gain scores on the widely used Test Of Astronomy STandards (TOAST). Usually used in a pre-test and post-test research design, one might naturally assume that the pre-course differences observed between high- and low-scoring college students might be due in large part to their pre-existing motivation, interest, experience in science, and attitudes about astronomy. To explore this notion, 11 non-science majoring undergraduates taking ASTRO 101 at west coast community colleges were interviewed in the first few weeks of the course to better understand students' pre-existing affect toward learning astronomy with an eye toward predicting student success. In answering this question, we hope to contribute to our understanding of the incoming knowledge of students taking undergraduate introductory astronomy classes, but also gain insight into how faculty can best meet those students' needs and assist them in achieving success. Perhaps surprisingly, there was only weak correlation between students' motivation toward learning astronomy and their pre-test scores. Instead, the most fruitful predictor of TOAST pre-test scores was the quantity of pre-existing, informal, self-directed astronomy learning experiences.

  17. Combining University Student Self-Regulated Learning Indicators and Engagement with Online Learning Events to Predict Academic Performance

    Science.gov (United States)

    Pardo, Abelardo; Han, Feifei; Ellis, Robert A.

    2017-01-01

    Self-regulated learning theories are used to understand the reasons for different levels of university student academic performance. Similarly, learning analytics research proposes the combination of detailed data traces derived from technology-mediated tasks with a variety of algorithms to predict student academic performance. The former approach…

  18. Predicting High School Student Use of Learning Strategies: The Role of Preferred Learning Styles and Classroom Climate

    Science.gov (United States)

    Cheema, Jehanzeb; Kitsantas, Anastasia

    2016-01-01

    This study investigated the predictiveness of preferred learning styles (competitive and cooperative) and classroom climate (teacher support and disciplinary climate) on learning strategy use in mathematics. The student survey part of the Programme for International Student Assessment 2003 comprising of 4633 US observations was used in a weighted…

  19. Alternative Testing Methods for Predicting Health Risk from Environmental Exposures

    Directory of Open Access Journals (Sweden)

    Annamaria Colacci

    2014-08-01

    Full Text Available Alternative methods to animal testing are considered as promising tools to support the prediction of toxicological risks from environmental exposure. Among the alternative testing methods, the cell transformation assay (CTA appears to be one of the most appropriate approaches to predict the carcinogenic properties of single chemicals, complex mixtures and environmental pollutants. The BALB/c 3T3 CTA shows a good degree of concordance with the in vivo rodent carcinogenesis tests. Whole-genome transcriptomic profiling is performed to identify genes that are transcriptionally regulated by different kinds of exposures. Its use in cell models representative of target organs may help in understanding the mode of action and predicting the risk for human health. Aiming at associating the environmental exposure to health-adverse outcomes, we used an integrated approach including the 3T3 CTA and transcriptomics on target cells, in order to evaluate the effects of airborne particulate matter (PM on toxicological complex endpoints. Organic extracts obtained from PM2.5 and PM1 samples were evaluated in the 3T3 CTA in order to identify effects possibly associated with different aerodynamic diameters or airborne chemical components. The effects of the PM2.5 extracts on human health were assessed by using whole-genome 44 K oligo-microarray slides. Statistical analysis by GeneSpring GX identified genes whose expression was modulated in response to the cell treatment. Then, modulated genes were associated with pathways, biological processes and diseases through an extensive biological analysis. Data derived from in vitro methods and omics techniques could be valuable for monitoring the exposure to toxicants, understanding the modes of action via exposure-associated gene expression patterns and to highlight the role of genes in key events related to adversity.

  20. Factors Motivating Individuals to Consider Genetic Testing for Type 2 Diabetes Risk Prediction.

    Directory of Open Access Journals (Sweden)

    Jennifer Wessel

    Full Text Available The purpose of this study was to identify attitudes and perceptions of willingness to participate in genetic testing for type 2 diabetes (T2D risk prediction in the general population. Adults (n = 598 were surveyed on attitudes about utilizing genetic testing to predict future risk of T2D. Participants were recruited from public libraries (53%, online registry (37% and a safety net hospital emergency department (10%. Respondents were 37 ± 11 years old, primarily White (54%, female (69%, college educated (46%, with an annual income ≥$25,000 (56%. Half of participants were interested in genetic testing for T2D (52% and 81% agreed/strongly agreed genetic testing should be available to the public. Only 57% of individuals knew T2D is preventable. A multivariate model to predict interest in genetic testing was adjusted for age, gender, recruitment location and BMI; significant predictors were motivation (high perceived personal risk of T2D [OR = 4.38 (1.76, 10.9]; family history [OR = 2.56 (1.46, 4.48]; desire to know risk prior to disease onset [OR = 3.25 (1.94, 5.42]; and knowing T2D is preventable [OR = 2.11 (1.24, 3.60], intention (if the cost is free [OR = 10.2 (4.27, 24.6]; and learning T2D is preventable [OR = 5.18 (1.95, 13.7] and trust of genetic testing results [OR = 0.03 (0.003, 0.30]. Individuals are interested in genetic testing for T2D risk which offers unique information that is personalized. Financial accessibility, validity of the test and availability of diabetes prevention programs were identified as predictors of interest in T2D testing.

  1. Development of Predictive QSAR Models of 4-Thiazolidinones Antitrypanosomal Activity using Modern Machine Learning Algorithms.

    Science.gov (United States)

    Kryshchyshyn, Anna; Devinyak, Oleg; Kaminskyy, Danylo; Grellier, Philippe; Lesyk, Roman

    2017-11-14

    This paper presents novel QSAR models for the prediction of antitrypanosomal activity among thiazolidines and related heterocycles. The performance of four machine learning algorithms: Random Forest regression, Stochastic gradient boosting, Multivariate adaptive regression splines and Gaussian processes regression have been studied in order to reach better levels of predictivity. The results for Random Forest and Gaussian processes regression are comparable and outperform other studied methods. The preliminary descriptor selection with Boruta method improved the outcome of machine learning methods. The two novel QSAR-models developed with Random Forest and Gaussian processes regression algorithms have good predictive ability, which was proved by the external evaluation of the test set with corresponding Q 2 ext =0.812 and Q 2 ext =0.830. The obtained models can be used further for in silico screening of virtual libraries in the same chemical domain in order to find new antitrypanosomal agents. Thorough analysis of descriptors influence in the QSAR models and interpretation of their chemical meaning allows to highlight a number of structure-activity relationships. The presence of phenyl rings with electron-withdrawing atoms or groups in para-position, increased number of aromatic rings, high branching but short chains, high HOMO energy, and the introduction of 1-substituted 2-indolyl fragment into the molecular structure have been recognized as trypanocidal activity prerequisites. © 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim.

  2. Social Emotional Learning and Educational Stress: A Predictive Model

    Science.gov (United States)

    Arslan, Serhat

    2015-01-01

    The purpose of this study is to examine the relationship between social emotional learning and educational stress. Participants were 321 elementary students. Social emotional learning and educational stress scale were used as measures. The relationships between social emotional learning and educational stress were examined using correlation…

  3. Different protein-protein interface patterns predicted by different machine learning methods.

    Science.gov (United States)

    Wang, Wei; Yang, Yongxiao; Yin, Jianxin; Gong, Xinqi

    2017-11-22

    Different types of protein-protein interactions make different protein-protein interface patterns. Different machine learning methods are suitable to deal with different types of data. Then, is it the same situation that different interface patterns are preferred for prediction by different machine learning methods? Here, four different machine learning methods were employed to predict protein-protein interface residue pairs on different interface patterns. The performances of the methods for different types of proteins are different, which suggest that different machine learning methods tend to predict different protein-protein interface patterns. We made use of ANOVA and variable selection to prove our result. Our proposed methods taking advantages of different single methods also got a good prediction result compared to single methods. In addition to the prediction of protein-protein interactions, this idea can be extended to other research areas such as protein structure prediction and design.

  4. Optimized Extreme Learning Machine for Power System Transient Stability Prediction Using Synchrophasors

    Directory of Open Access Journals (Sweden)

    Yanjun Zhang

    2015-01-01

    Full Text Available A new optimized extreme learning machine- (ELM- based method for power system transient stability prediction (TSP using synchrophasors is presented in this paper. First, the input features symbolizing the transient stability of power systems are extracted from synchronized measurements. Then, an ELM classifier is employed to build the TSP model. And finally, the optimal parameters of the model are optimized by using the improved particle swarm optimization (IPSO algorithm. The novelty of the proposal is in the fact that it improves the prediction performance of the ELM-based TSP model by using IPSO to optimize the parameters of the model with synchrophasors. And finally, based on the test results on both IEEE 39-bus system and a large-scale real power system, the correctness and validity of the presented approach are verified.

  5. Nonparametric predictive inference for combining diagnostic tests with parametric copula

    Science.gov (United States)

    Muhammad, Noryanti; Coolen, F. P. A.; Coolen-Maturi, T.

    2017-09-01

    Measuring the accuracy of diagnostic tests is crucial in many application areas including medicine and health care. The Receiver Operating Characteristic (ROC) curve is a popular statistical tool for describing the performance of diagnostic tests. The area under the ROC curve (AUC) is often used as a measure of the overall performance of the diagnostic test. In this paper, we interest in developing strategies for combining test results in order to increase the diagnostic accuracy. We introduce nonparametric predictive inference (NPI) for combining two diagnostic test results with considering dependence structure using parametric copula. NPI is a frequentist statistical framework for inference on a future observation based on past data observations. NPI uses lower and upper probabilities to quantify uncertainty and is based on only a few modelling assumptions. While copula is a well-known statistical concept for modelling dependence of random variables. A copula is a joint distribution function whose marginals are all uniformly distributed and it can be used to model the dependence separately from the marginal distributions. In this research, we estimate the copula density using a parametric method which is maximum likelihood estimator (MLE). We investigate the performance of this proposed method via data sets from the literature and discuss results to show how our method performs for different family of copulas. Finally, we briefly outline related challenges and opportunities for future research.

  6. How long will my mouse live? Machine learning approaches for prediction of mouse life span.

    Science.gov (United States)

    Swindell, William R; Harper, James M; Miller, Richard A

    2008-09-01

    Prediction of individual life span based on characteristics evaluated at middle-age represents a challenging objective for aging research. In this study, we used machine learning algorithms to construct models that predict life span in a stock of genetically heterogeneous mice. Life-span prediction accuracy of 22 algorithms was evaluated using a cross-validation approach, in which models were trained and tested with distinct subsets of data. Using a combination of body weight and T-cell subset measures evaluated before 2 years of age, we show that the life-span quartile to which an individual mouse belongs can be predicted with an accuracy of 35.3% (+/-0.10%). This result provides a new benchmark for the development of life-span-predictive models, but improvement can be expected through identification of new predictor variables and development of computational approaches. Future work in this direction can provide tools for aging research and will shed light on associations between phenotypic traits and longevity.

  7. A New Tool for CME Arrival Time Prediction using Machine Learning Algorithms: CAT-PUMA

    Science.gov (United States)

    Liu, Jiajia; Ye, Yudong; Shen, Chenglong; Wang, Yuming; Erdélyi, Robert

    2018-03-01

    Coronal mass ejections (CMEs) are arguably the most violent eruptions in the solar system. CMEs can cause severe disturbances in interplanetary space and can even affect human activities in many aspects, causing damage to infrastructure and loss of revenue. Fast and accurate prediction of CME arrival time is vital to minimize the disruption that CMEs may cause when interacting with geospace. In this paper, we propose a new approach for partial-/full halo CME Arrival Time Prediction Using Machine learning Algorithms (CAT-PUMA). Via detailed analysis of the CME features and solar-wind parameters, we build a prediction engine taking advantage of 182 previously observed geo-effective partial-/full halo CMEs and using algorithms of the Support Vector Machine. We demonstrate that CAT-PUMA is accurate and fast. In particular, predictions made after applying CAT-PUMA to a test set unknown to the engine show a mean absolute prediction error of ∼5.9 hr within the CME arrival time, with 54% of the predictions having absolute errors less than 5.9 hr. Comparisons with other models reveal that CAT-PUMA has a more accurate prediction for 77% of the events investigated that can be carried out very quickly, i.e., within minutes of providing the necessary input parameters of a CME. A practical guide containing the CAT-PUMA engine and the source code of two examples are available in the Appendix, allowing the community to perform their own applications for prediction using CAT-PUMA.

  8. A Supervised Learning Process to Validate Online Disease Reports for Use in Predictive Models.

    Science.gov (United States)

    Patching, Helena M M; Hudson, Laurence M; Cooke, Warrick; Garcia, Andres J; Hay, Simon I; Roberts, Mark; Moyes, Catherine L

    2015-12-01

    Pathogen distribution models that predict spatial variation in disease occurrence require data from a large number of geographic locations to generate disease risk maps. Traditionally, this process has used data from public health reporting systems; however, using online reports of new infections could speed up the process dramatically. Data from both public health systems and online sources must be validated before they can be used, but no mechanisms exist to validate data from online media reports. We have developed a supervised learning process to validate geolocated disease outbreak data in a timely manner. The process uses three input features, the data source and two metrics derived from the location of each disease occurrence. The location of disease occurrence provides information on the probability of disease occurrence at that location based on environmental and socioeconomic factors and the distance within or outside the current known disease extent. The process also uses validation scores, generated by disease experts who review a subset of the data, to build a training data set. The aim of the supervised learning process is to generate validation scores that can be used as weights going into the pathogen distribution model. After analyzing the three input features and testing the performance of alternative processes, we selected a cascade of ensembles comprising logistic regressors. Parameter values for the training data subset size, number of predictors, and number of layers in the cascade were tested before the process was deployed. The final configuration was tested using data for two contrasting diseases (dengue and cholera), and 66%-79% of data points were assigned a validation score. The remaining data points are scored by the experts, and the results inform the training data set for the next set of predictors, as well as going to the pathogen distribution model. The new supervised learning process has been implemented within our live site and is

  9. Unique and shared validity of the "Wechsler logical memory test", the "California verbal learning test", and the "verbal learning and memory test" in patients with epilepsy.

    Science.gov (United States)

    Helmstaedter, Christoph; Wietzke, Jennifer; Lutz, Martin T

    2009-12-01

    This study was set-up to evaluate the construct validity of three verbal memory tests in epilepsy patients. Sixty-one consecutively evaluated patients with temporal lobe epilepsy (TLE) or extra-temporal epilepsy (E-TLE) underwent testing with the verbal learning and memory test (VLMT, the German equivalent of the Rey auditory verbal learning test, RAVLT); the California verbal learning test (CVLT); the logical memory and digit span subtests of the Wechsler memory scale, revised (WMS-R); and testing of intelligence, attention, speech and executive functions. Factor analysis of the memory tests resulted in test-specific rather than test over-spanning factors. Parameters of the CVLT and WMS-R, and to a much lesser degree of the VLMT, were highly correlated with attention, language function and vocabulary. Delayed recall measures of logical memory and the VLMT differentiated TLE from E-TLE. Learning and memory scores off all three tests differentiated mesial temporal sclerosis from other pathologies. A lateralization of the epilepsy was possible only for a subsample of 15 patients with mesial TLE. Although the three tests provide overlapping indicators for a temporal lobe epilepsy or a mesial pathology, they can hardly be taken in exchange. The tests have different demands on semantic processing and memory organization, and they appear differentially sensitive to performance in non-memory domains. The tests capability to lateralize appears to be poor. The findings encourage the further discussion of the dependency of memory outcomes on test selection.

  10. The Enhancement of Communication Skill and Prediction Skill in Colloidal Concept by Problem Solving Learning

    OpenAIRE

    Anggraini, Agita Dzulhajh; Fadiawati, Noor; Diawati, Chansyanah

    2012-01-01

    Accuracy educators in selecting and implementing learning models influence students' science process skills. Models of learning that can be applied to improve science process skills and tend constructivist among athers learning model of problem solving. This research was conducted to describe the effectiveness of the learning model of problem solving in improving communication skills and prediction skills. Subjects in this research were students of high school YP Unila Bandar Lampung Even ...

  11. In Silico Prediction of Chemicals Binding to Aromatase with Machine Learning Methods.

    Science.gov (United States)

    Du, Hanwen; Cai, Yingchun; Yang, Hongbin; Zhang, Hongxiao; Xue, Yuhan; Liu, Guixia; Tang, Yun; Li, Weihua

    2017-05-15

    Environmental chemicals may affect endocrine systems through multiple mechanisms, one of which is via effects on aromatase (also known as CYP19A1), an enzyme critical for maintaining the normal balance of estrogens and androgens in the body. Therefore, rapid and efficient identification of aromatase-related endocrine disrupting chemicals (EDCs) is important for toxicology and environment risk assessment. In this study, on the basis of the Tox21 10K compound library, in silico classification models for predicting aromatase binders/nonbinders were constructed by machine learning methods. To improve the prediction ability of the models, a combined classifier (CC) strategy that combines different independent machine learning methods was adopted. Performances of the models were measured by test and external validation sets containing 1336 and 216 chemicals, respectively. The best model was obtained with the MACCS (Molecular Access System) fingerprint and CC method, which exhibited an accuracy of 0.84 for the test set and 0.91 for the external validation set. Additionally, several representative substructures for characterizing aromatase binders, such as ketone, lactone, and nitrogen-containing derivatives, were identified using information gain and substructure frequency analysis. Our study provided a systematic assessment of chemicals binding to aromatase. The built models can be helpful to rapidly identify potential EDCs targeting aromatase.

  12. Evaluation of a deep learning architecture for MR imaging prediction of ATRX in glioma patients

    Science.gov (United States)

    Korfiatis, Panagiotis; Kline, Timothy L.; Erickson, Bradley J.

    2018-02-01

    Predicting mutation/loss of alpha-thalassemia/mental retardation syndrome X-linked (ATRX) gene utilizing MR imaging is of high importance since it is a predictor of response and prognosis in brain tumors. In this study, we compare a deep neural network approach based on a residual deep neural network (ResNet) architecture and one based on a classical machine learning approach and evaluate their ability in predicting ATRX mutation status without the need for a distinct tumor segmentation step. We found that the ResNet50 (50 layers) architecture, pre trained on ImageNet data was the best performing model, achieving an accuracy of 0.91 for the test set (classification of a slice as no tumor, ATRX mutated, or mutated) in terms of f1 score in a test set of 35 cases. The SVM classifier achieved 0.63 for differentiating the Flair signal abnormality regions from the test patients based on their mutation status. We report a method that alleviates the need for extensive preprocessing and acts as a proof of concept that deep neural network architectures can be used to predict molecular biomarkers from routine medical images.

  13. Prediction of safety critical software operational reliability from test reliability using testing environment factors

    International Nuclear Information System (INIS)

    Jung, Hoan Sung; Seong, Poong Hyun

    1999-01-01

    It has been a critical issue to predict the safety critical software reliability in nuclear engineering area. For many years, many researches have focused on the quantification of software reliability and there have been many models developed to quantify software reliability. Most software reliability models estimate the reliability with the failure data collected during the test assuming that the test environments well represent the operation profile. User's interest is however on the operational reliability rather than on the test reliability. The experiences show that the operational reliability is higher than the test reliability. With the assumption that the difference in reliability results from the change of environment, from testing to operation, testing environment factors comprising the aging factor and the coverage factor are developed in this paper and used to predict the ultimate operational reliability with the failure data in testing phase. It is by incorporating test environments applied beyond the operational profile into testing environment factors. The application results show that the proposed method can estimate the operational reliability accurately. (Author). 14 refs., 1 tab., 1 fig

  14. Long-term associative learning predicts verbal short-term memory performance.

    Science.gov (United States)

    Jones, Gary; Macken, Bill

    2018-02-01

    Studies using tests such as digit span and nonword repetition have implicated short-term memory across a range of developmental domains. Such tests ostensibly assess specialized processes for the short-term manipulation and maintenance of information that are often argued to enable long-term learning. However, there is considerable evidence for an influence of long-term linguistic learning on performance in short-term memory tasks that brings into question the role of a specialized short-term memory system separate from long-term knowledge. Using natural language corpora, we show experimentally and computationally that performance on three widely used measures of short-term memory (digit span, nonword repetition, and sentence recall) can be predicted from simple associative learning operating on the linguistic environment to which a typical child may have been exposed. The findings support the broad view that short-term verbal memory performance reflects the application of long-term language knowledge to the experimental setting.

  15. Visuospatial Aptitude Testing Differentially Predicts Simulated Surgical Skill.

    Science.gov (United States)

    Hinchcliff, Emily; Green, Isabel; Destephano, Christopher; Cox, Mary; Smink, Douglas; Kumar, Amanika; Hokenstad, Erik; Bengtson, Joan; Cohen, Sarah

    2018-02-05

    To determine if visuospatial perception (VSP) testing is correlated to simulated or intraoperative surgical performance as rated by the American College of Graduate Medical Education (ACGME) milestones. Classification II-2 SETTING: Two academic training institutions PARTICIPANTS: 41 residents, including 19 Brigham and Women's Hospital and 22 Mayo Clinic residents from three different specialties (OBGYN, general surgery, urology). Participants underwent three different tests: visuospatial perception testing (VSP), Fundamentals of Laparoscopic Surgery (FLS®) peg transfer, and DaVinci robotic simulation peg transfer. Surgical grading from the ACGME milestones tool was obtained for each participant. Demographic and subject background information was also collected including specialty, year of training, prior experience with simulated skills, and surgical interest. Standard statistical analysis using Student's t test were performed, and correlations were determined using adjusted linear regression models. In univariate analysis, BWH and Mayo training programs differed in both times and overall scores for both FLS® peg transfer and DaVinci robotic simulation peg transfer (p<0.05 for all). Additionally, type of residency training impacted time and overall score on robotic peg transfer. Familiarity with tasks correlated with higher score and faster task completion (p= 0.05 for all except VSP score). There was no difference in VSP scores by program, specialty, or year of training. In adjusted linear regression modeling, VSP testing was correlated only to robotic peg transfer skills (average time p=0.006, overall score p=0.001). Milestones did not correlate to either VSP or surgical simulation testing. VSP score was correlated with robotic simulation skills but not with FLS skills or ACGME milestones. This suggests that the ability of VSP score to predict competence differs between tasks. Therefore, further investigation is required into aptitude testing, especially prior

  16. Void fraction prediction of NUPEC PSBT tests by CATHARE code

    International Nuclear Information System (INIS)

    Del Nevo, A.; Michelotti, L.; Moretti, F.; Rozzia, D.; D'Auria, F.

    2011-01-01

    The current generation of thermal-hydraulic system codes benefits of about sixty years of experiments and forty years of development and are considered mature tools to provide best estimate description of phenomena and detailed reactor system representations. However, there are continuous needs for checking the code capabilities in representing nuclear system, for drawing attention to their weak points, for identifying models which need to be refined for best-estimate calculations. Prediction of void fraction and Departure from Nucleate Boiling (DNB) in system thermal-hydraulics is currently based on empirical approaches. The database carried out by Nuclear Power Engineering Corporation (NUPEC), Japan addresses these issues. It is suitable for supporting the development of new computational tools based on more mechanistic approaches (i.e. three-field codes, two-phase CFD, etc.) as well as for validating current generation of thermal-hydraulic system codes. Selected experiments belonging to this database are used for the OECD/NRC PSBT benchmark. The paper reviews the activity carried out by CATHARE2 code on the basis of the subchannel (four test sections) and presents rod bundle (different axial power profile and test sections) experiments available in the database in steady state and transient conditions. The results demonstrate the accuracy of the code in predicting the void fraction in different thermal-hydraulic conditions. The tests are performed varying the pressure, coolant temperature, mass flow and power. Sensitivity analyses are carried out addressing nodalization effect and the influence of the initial and boundary conditions of the tests. (author)

  17. Consumer preferences for the predictive genetic test for Alzheimer disease.

    Science.gov (United States)

    Huang, Ming-Yi; Huston, Sally A; Perri, Matthew

    2014-04-01

    The purpose of this study was to assess consumer preferences for predictive genetic testing for Alzheimer disease in the United States. A rating conjoint analysis was conducted using an anonymous online survey distributed by Qualtrics to a general population panel in April 2011 in the United States. The study design included three attributes: Accuracy (40%, 80%, and 100%), Treatment Availability (Cure is available/Drug for symptom relief but no cure), and Anonymity (Anonymous/Not anonymous). A total of 12 scenarios were used to elicit people's preference, assessed by an 11-point scale. The respondents also indicated their highest willingness-to-pay (WTP) for each scenario through open-ended questions. A total of 295 responses were collected over 4 days. The most important attribute for the aggregate model was Accuracy, contributing 64.73% to the preference rating. Treatment Availability and Anonymity contributed 20.72% and 14.59%, respectively, to the preference rating. The median WTP for the highest-rating scenario (Accuracy 100%, a cure is available, test result is anonymous) was $100 (mean = $276). The median WTP for the lowest-rating scenario (40% accuracy, no cure but drugs for symptom relief, not anonymous) was zero (mean = $34). The results of this study highlight attributes people find important when making the hypothetical decision to obtain an AD genetic test. These results should be of interests to policy makers, genetic test developers and health care providers.

  18. CAN UPPER EXTREMITY FUNCTIONAL TESTS PREDICT THE SOFTBALL THROW FOR DISTANCE: A PREDICTIVE VALIDITY INVESTIGATION

    Science.gov (United States)

    Hanney, William J.; Kolber, Morey J.; Davies, George J.; Riemann, Bryan

    2011-01-01

    Introduction: Understanding the relationships between performance tests and sport activity is important to the rehabilitation specialist. The purpose of this study was two- fold: 1) To identify if relationships exist between tests of upper body strength and power (Single Arm Seated Shot Put, Timed Push-Up, Timed Modified Pull-Up, and The Davies Closed Kinetic Chain Upper Extremity Stability Test, and the softball throw for distance), 2) To determine which variable or group of variables best predicts the performance of a sport specific task (the softball throw for distance). Methods: One hundred eighty subjects (111 females and 69 males, aged 18-45 years) performed the 5 upper extremity tests. The Pearson product moment correlation and a stepwise regression were used to determine whether relationships existed between performance on the tests and which upper extremity test result best explained the performance on the softball throw for distance. Results: There were significant correlations (r=.33 to r=.70, p=0.001) between performance on all of the tests. The modified pull-up test was the best predictor of the performance on the softball throw for distance (r2= 48.7), explaining 48.7% of variation in performance. When weight, height, and age were added to the regression equation the r2 values increased to 64.5, 66.2, and 67.5 respectively. Conclusion: The results of this study indicate that several upper extremity tests demonstrate significant relationships with one another and with the softball throw for distance. The modified pull up test was the best predictor of performance on the softball throw for distance. PMID:21712942

  19. Prediction of Human Drug Targets and Their Interactions Using Machine Learning Methods: Current and Future Perspectives.

    Science.gov (United States)

    Nath, Abhigyan; Kumari, Priyanka; Chaube, Radha

    2018-01-01

    Identification of drug targets and drug target interactions are important steps in the drug-discovery pipeline. Successful computational prediction methods can reduce the cost and time demanded by the experimental methods. Knowledge of putative drug targets and their interactions can be very useful for drug repurposing. Supervised machine learning methods have been very useful in drug target prediction and in prediction of drug target interactions. Here, we describe the details for developing prediction models using supervised learning techniques for human drug target prediction and their interactions.

  20. Applying machine learning to predict patient-specific current CD 4 ...

    African Journals Online (AJOL)

    This work shows the application of machine learning to predict current CD4 cell count of an HIV-positive patient using genome sequences, viral load and time. A regression model predicting actual CD4 cell counts and a classification model predicting if a patient's CD4 cell count is less than 200 was built using a support ...

  1. An analysis of a digital variant of the Trail Making Test using machine learning techniques.

    Science.gov (United States)

    Dahmen, Jessamyn; Cook, Diane; Fellows, Robert; Schmitter-Edgecombe, Maureen

    2017-01-01

    The goal of this work is to develop a digital version of a standard cognitive assessment, the Trail Making Test (TMT), and assess its utility. This paper introduces a novel digital version of the TMT and introduces a machine learning based approach to assess its capabilities. Using digital Trail Making Test (dTMT) data collected from (N = 54) older adult participants as feature sets, we use machine learning techniques to analyze the utility of the dTMT and evaluate the insights provided by the digital features. Predicted TMT scores correlate well with clinical digital test scores (r = 0.98) and paper time to completion scores (r = 0.65). Predicted TICS exhibited a small correlation with clinically derived TICS scores (r = 0.12 Part A, r = 0.10 Part B). Predicted FAB scores exhibited a small correlation with clinically derived FAB scores (r = 0.13 Part A, r = 0.29 for Part B). Digitally derived features were also used to predict diagnosis (AUC of 0.65). Our findings indicate that the dTMT is capable of measuring the same aspects of cognition as the paper-based TMT. Furthermore, the dTMT's additional data may be able to help monitor other cognitive processes not captured by the paper-based TMT alone.

  2. A Machine Learning Approach to Test Data Generation

    DEFF Research Database (Denmark)

    Christiansen, Henning; Dahmcke, Christina Mackeprang

    2007-01-01

    been tested, and a more thorough statistical foundation is required. We propose to use logic-statistical modelling methods for machine-learning for analyzing existing and manually marked up data, integrated with the generation of new, artificial data. More specifically, we suggest to use the PRISM...... system developed by Sato and Kameya. Based on logic programming extended with random variables and parameter learning, PRISM appears as a powerful modelling environment, which subsumes HMMs and a wide range of other methods, all embedded in a declarative language. We illustrate these principles here...

  3. A comparison of the efficacy of test-driven learning versus self-assessment learning

    Science.gov (United States)

    He, Xiaohua; Canty, Anne

    2013-01-01

    Objective We compared self-assessment and test-driven learning in two groups of students who studied the same subject. Methods This was a randomized comparative experimental study. The subjects were 259 first-quarter students who were divided into a test group and a self-assessment group based on the methods they used for their learning assessments. We measured the scores and difficulty levels of 3 formal written exams. Students' attitudes toward self-assessment or test-driven learning were surveyed. Results The mean scores of exam 1, exam 2, and a summative exam were 34 (±6), 32 (±8), and 44 (±6) for the self-assessment group, respectively, with corresponding scores of 33 (±6), 33 (±7), 43 (±6) for the test group. There were no significant differences in the mean scores on all 3 tests between the two groups (p > .05). Of the students in the self-assessment group, 64% scored at least 90%, whereas 47% of students in the test group answered at least 90% of the questions correctly (p self-assessment and tests could have a significant impact on students' learning, but each offers different strengths and weaknesses. PMID:23957317

  4. Prediction of the Chloride Resistance of Concrete Modified with High Calcium Fly Ash Using Machine Learning.

    Science.gov (United States)

    Marks, Michał; Glinicki, Michał A; Gibas, Karolina

    2015-12-11

    The aim of the study was to generate rules for the prediction of the chloride resistance of concrete modified with high calcium fly ash using machine learning methods. The rapid chloride permeability test, according to the Nordtest Method Build 492, was used for determining the chloride ions' penetration in concrete containing high calcium fly ash (HCFA) for partial replacement of Portland cement. The results of the performed tests were used as the training set to generate rules describing the relation between material composition and the chloride resistance. Multiple methods for rule generation were applied and compared. The rules generated by algorithm J48 from the Weka workbench provided the means for adequate classification of plain concretes and concretes modified with high calcium fly ash as materials of good, acceptable or unacceptable resistance to chloride penetration.

  5. Machine Learning Methods for Prediction of CDK-Inhibitors

    Science.gov (United States)

    Ramana, Jayashree; Gupta, Dinesh

    2010-01-01

    Progression through the cell cycle involves the coordinated activities of a suite of cyclin/cyclin-dependent kinase (CDK) complexes. The activities of the complexes are regulated by CDK inhibitors (CDKIs). Apart from its role as cell cycle regulators, CDKIs are involved in apoptosis, transcriptional regulation, cell fate determination, cell migration and cytoskeletal dynamics. As the complexes perform crucial and diverse functions, these are important drug targets for tumour and stem cell therapeutic interventions. However, CDKIs are represented by proteins with considerable sequence heterogeneity and may fail to be identified by simple similarity search methods. In this work we have evaluated and developed machine learning methods for identification of CDKIs. We used different compositional features and evolutionary information in the form of PSSMs, from CDKIs and non-CDKIs for generating SVM and ANN classifiers. In the first stage, both the ANN and SVM models were evaluated using Leave-One-Out Cross-Validation and in the second stage these were tested on independent data sets. The PSSM-based SVM model emerged as the best classifier in both the stages and is publicly available through a user-friendly web interface at http://bioinfo.icgeb.res.in/cdkipred. PMID:20967128

  6. Implementation of Chaotic Gaussian Particle Swarm Optimization for Optimize Learning-to-Rank Software Defect Prediction Model Construction

    Science.gov (United States)

    Buchari, M. A.; Mardiyanto, S.; Hendradjaya, B.

    2018-03-01

    Finding the existence of software defect as early as possible is the purpose of research about software defect prediction. Software defect prediction activity is required to not only state the existence of defects, but also to be able to give a list of priorities which modules require a more intensive test. Therefore, the allocation of test resources can be managed efficiently. Learning to rank is one of the approach that can provide defect module ranking data for the purposes of software testing. In this study, we propose a meta-heuristic chaotic Gaussian particle swarm optimization to improve the accuracy of learning to rank software defect prediction approach. We have used 11 public benchmark data sets as experimental data. Our overall results has demonstrated that the prediction models construct using Chaotic Gaussian Particle Swarm Optimization gets better accuracy on 5 data sets, ties in 5 data sets and gets worse in 1 data sets. Thus, we conclude that the application of Chaotic Gaussian Particle Swarm Optimization in Learning-to-Rank approach can improve the accuracy of the defect module ranking in data sets that have high-dimensional features.

  7. Predictive coding accelerates word recognition and learning in the early stages of language development.

    Science.gov (United States)

    Ylinen, Sari; Bosseler, Alexis; Junttila, Katja; Huotilainen, Minna

    2017-11-01

    The ability to predict future events in the environment and learn from them is a fundamental component of adaptive behavior across species. Here we propose that inferring predictions facilitates speech processing and word learning in the early stages of language development. Twelve- and 24-month olds' electrophysiological brain responses to heard syllables are faster and more robust when the preceding word context predicts the ending of a familiar word. For unfamiliar, novel word forms, however, word-expectancy violation generates a prediction error response, the strength of which significantly correlates with children's vocabulary scores at 12 months. These results suggest that predictive coding may accelerate word recognition and support early learning of novel words, including not only the learning of heard word forms but also their mapping to meanings. Prediction error may mediate learning via attention, since infants' attention allocation to the entire learning situation in natural environments could account for the link between prediction error and the understanding of word meanings. On the whole, the present results on predictive coding support the view that principles of brain function reported across domains in humans and non-human animals apply to language and its development in the infant brain. A video abstract of this article can be viewed at: http://hy.fi/unitube/video/e1cbb495-41d8-462e-8660-0864a1abd02c. [Correction added on 27 January 2017, after first online publication: The video abstract link was added.]. © 2016 John Wiley & Sons Ltd.

  8. Predicting Pilot Behavior in Medium Scale Scenarios Using Game Theory and Reinforcement Learning

    Science.gov (United States)

    Yildiz, Yildiray; Agogino, Adrian; Brat, Guillaume

    2013-01-01

    Effective automation is critical in achieving the capacity and safety goals of the Next Generation Air Traffic System. Unfortunately creating integration and validation tools for such automation is difficult as the interactions between automation and their human counterparts is complex and unpredictable. This validation becomes even more difficult as we integrate wide-reaching technologies that affect the behavior of different decision makers in the system such as pilots, controllers and airlines. While overt short-term behavior changes can be explicitly modeled with traditional agent modeling systems, subtle behavior changes caused by the integration of new technologies may snowball into larger problems and be very hard to detect. To overcome these obstacles, we show how integration of new technologies can be validated by learning behavior models based on goals. In this framework, human participants are not modeled explicitly. Instead, their goals are modeled and through reinforcement learning their actions are predicted. The main advantage to this approach is that modeling is done within the context of the entire system allowing for accurate modeling of all participants as they interact as a whole. In addition such an approach allows for efficient trade studies and feasibility testing on a wide range of automation scenarios. The goal of this paper is to test that such an approach is feasible. To do this we implement this approach using a simple discrete-state learning system on a scenario where 50 aircraft need to self-navigate using Automatic Dependent Surveillance-Broadcast (ADS-B) information. In this scenario, we show how the approach can be used to predict the ability of pilots to adequately balance aircraft separation and fly efficient paths. We present results with several levels of complexity and airspace congestion.

  9. An ensemble machine learning approach to predict survival in breast cancer.

    Science.gov (United States)

    Djebbari, Amira; Liu, Ziying; Phan, Sieu; Famili, Fazel

    2008-01-01

    Current breast cancer predictive signatures are not unique. Can we use this fact to our advantage to improve prediction? From the machine learning perspective, it is well known that combining multiple classifiers can improve classification performance. We propose an ensemble machine learning approach which consists of choosing feature subsets and learning predictive models from them. We then combine models based on certain model fusion criteria and we also introduce a tuning parameter to control sensitivity. Our method significantly improves classification performance with a particular emphasis on sensitivity which is critical to avoid misclassifying poor prognosis patients as good prognosis.

  10. Transfer of test-enhanced learning: Meta-analytic review and synthesis.

    Science.gov (United States)

    Pan, Steven C; Rickard, Timothy C

    2018-05-07

    Attempting recall of information from memory, as occurs when taking a practice test, is one of the most potent training techniques known to learning science. However, does testing yield learning that transfers to different contexts? In the present article, we report the findings of the first comprehensive meta-analytic review into that question. Our review encompassed 192 transfer effect sizes extracted from 122 experiments and 67 published and unpublished articles (N = 10,382) that together comprise more than 40 years of research. A random-effects model revealed that testing can yield transferrable learning as measured relative to a nontesting reexposure control condition (d = 0.40, 95% CI [0.31, 0.50]). That transfer of learning is greatest across test formats, to application and inference questions, to problems involving medical diagnoses, and to mediator and related word cues; it is weakest to rearranged stimulus-response items, to untested materials seen during initial study, and to problems involving worked examples. Moderator analyses further indicated that response congruency and elaborated retrieval practice, as well as initial test performance, strongly influence the likelihood of positive transfer. In two assessments for publication bias using PET-PEESE and various selection methods, the moderator effect sizes were minimally affected. However, the intercept predictions were substantially reduced, often indicating no positive transfer when none of the aforementioned moderators are present. Overall, our results motivate a three-factor framework for transfer of test-enhanced learning and have practical implications for the effective use of practice testing in educational and other training contexts. (PsycINFO Database Record (c) 2018 APA, all rights reserved).

  11. Parent Ratings of Impulsivity and Inhibition Predict State Testing Scores

    Directory of Open Access Journals (Sweden)

    Rebecca A. Lundwall

    2018-03-01

    Full Text Available One principle of cognitive development is that earlier intervention for educational difficulties tends to improve outcomes such as future educational and career success. One possible way to help students who struggle is to determine if they process information differently. Such determination might lead to clues for interventions. For example, early information processing requires attention before the information can be identified, encoded, and stored. The aim of the present study was to investigate whether parent ratings of inattention, inhibition, and impulsivity, and whether error rate on a reflexive attention task could be used to predict child scores on state standardized tests. Finding such an association could provide assistance to educators in identifying academically struggling children who might require targeted educational interventions. Children (N = 203 were invited to complete a peripheral cueing task (which measures the automatic reorienting of the brain’s attentional resources from one location to another. While the children completed the task, their parents completed a questionnaire. The questionnaire gathered information on broad indicators of child functioning, including observable behaviors of impulsivity, inattention, and inhibition, as well as state academic scores (which the parent retrieved online from their school. We used sequential regression to analyze contributions of error rate and parent-rated behaviors in predicting six academic scores. In one of the six analyses (for science, we found that the improvement was significant from the simplified model (with only family income, child age, and sex as predictors to the full model (adding error rate and three parent-rated behaviors. Two additional analyses (reading and social studies showed near significant improvement from simplified to full models. Parent-rated behaviors were significant predictors in all three of these analyses. In the reading score analysis

  12. Ophthalmodynamometry for ICP prediction and pilot test on Mt. Everest.

    Science.gov (United States)

    Querfurth, Henry W; Lieberman, Philip; Arms, Steve; Mundell, Steve; Bennett, Michael; van Horne, Craig

    2010-11-01

    A recent development in non-invasive techniques to predict intracranial pressure (ICP) termed venous ophthalmodynamometry (vODM) has made measurements in absolute units possible. However, there has been little progress to show utility in the clinic or field. One important application would be to predict changes in actual ICP during adaptive responses to physiologic stress such as hypoxia. A causal relationship between raised intracranial pressure and acute mountain sickness (AMS) is suspected. Several MRI studies report that modest physiologic increases in cerebral volume, from swelling, normally accompany subacute ascent to simulated high altitudes. 1) Validate and calibrate an advanced, portable vODM instrument on intensive patients with raised intracranial pressure and 2) make pilot, non-invasive ICP estimations of normal subjects at increasing altitudes. The vODM was calibrated against actual ICP in 12 neurosurgical patients, most affected with acute hydrocephalus and monitored using ventriculostomy/pressure transducers. The operator was blinded to the transducer read-out. A clinical field test was then conducted on a variable data set of 42 volunteer trekkers and climbers scaling Mt. Everest, Nepal. Mean ICPs were estimated at several altitudes on the ascent both across and within subjects. Portable vODM measurements increased directly and linearly with ICP resulting in good predictability (r = 0.85). We also found that estimated ICP increases normally with altitude (10 ± 3 mm Hg; sea level to 20 ± 2 mm Hg; 6553 m) and that AMS symptoms did not correlate with raised ICP. vODM technology has potential to reliably estimate absolute ICP and is portable. Physiologic increases in ICP and mild-mod AMS are separate responses to high altitude, possibly reflecting swelling and vasoactive instability, respectively.

  13. Ophthalmodynamometry for ICP prediction and pilot test on Mt. Everest

    Directory of Open Access Journals (Sweden)

    Bennett Michael

    2010-11-01

    Full Text Available Abstract Background A recent development in non-invasive techniques to predict intracranial pressure (ICP termed venous ophthalmodynamometry (vODM has made measurements in absolute units possible. However, there has been little progress to show utility in the clinic or field. One important application would be to predict changes in actual ICP during adaptive responses to physiologic stress such as hypoxia. A causal relationship between raised intracranial pressure and acute mountain sickness (AMS is suspected. Several MRI studies report that modest physiologic increases in cerebral volume, from swelling, normally accompany subacute ascent to simulated high altitudes. Objectives 1 Validate and calibrate an advanced, portable vODM instrument on intensive patients with raised intracranial pressure and 2 make pilot, non-invasive ICP estimations of normal subjects at increasing altitudes. Methods The vODM was calibrated against actual ICP in 12 neurosurgical patients, most affected with acute hydrocephalus and monitored using ventriculostomy/pressure transducers. The operator was blinded to the transducer read-out. A clinical field test was then conducted on a variable data set of 42 volunteer trekkers and climbers scaling Mt. Everest, Nepal. Mean ICPs were estimated at several altitudes on the ascent both across and within subjects. Results Portable vODM measurements increased directly and linearly with ICP resulting in good predictability (r = 0.85. We also found that estimated ICP increases normally with altitude (10 ± 3 mm Hg; sea level to 20 ± 2 mm Hg; 6553 m and that AMS symptoms did not correlate with raised ICP. Conclusion vODM technology has potential to reliably estimate absolute ICP and is portable. Physiologic increases in ICP and mild-mod AMS are separate responses to high altitude, possibly reflecting swelling and vasoactive instability, respectively.

  14. Teacher response to learning disability: a test of attributional principles.

    Science.gov (United States)

    Clark, M D

    1997-01-01

    Attribution research has identified student ability and effort expended as causes of achievement outcomes that result in differing teacher affect, evaluative feedback, and expectation of future performance. Ninety-seven elementary-school general education teachers (84 women and 13 men) rated their responses to the test failures of hypothetical boys with and without learning disabilities. In most cases, greater reward and less punishment, less anger and more pity, and higher expectations of future failure followed the negative outcomes of the boys with learning disabilities, when compared with their nondisabled ability and effort matches, indicating that learning disability acts as a cause of achievement outcomes in the same way as ability and effort. This pattern of teacher affect and response can send negative messages that are often interpreted as low-ability cues, thus affecting students' self-esteem, sense of competence as learners, and motivation to achieve.

  15. Is Romantic Desire Predictable? Machine Learning Applied to Initial Romantic Attraction.

    Science.gov (United States)

    Joel, Samantha; Eastwick, Paul W; Finkel, Eli J

    2017-10-01

    Matchmaking companies and theoretical perspectives on close relationships suggest that initial attraction is, to some extent, a product of two people's self-reported traits and preferences. We used machine learning to test how well such measures predict people's overall tendencies to romantically desire other people (actor variance) and to be desired by other people (partner variance), as well as people's desire for specific partners above and beyond actor and partner variance (relationship variance). In two speed-dating studies, romantically unattached individuals completed more than 100 self-report measures about traits and preferences that past researchers have identified as being relevant to mate selection. Each participant met each opposite-sex participant attending a speed-dating event for a 4-min speed date. Random forests models predicted 4% to 18% of actor variance and 7% to 27% of partner variance; crucially, however, they were unable to predict relationship variance using any combination of traits and preferences reported before the dates. These results suggest that compatibility elements of human mating are challenging to predict before two people meet.

  16. eLearning Mobile App for Android and Ios "English Grammar Learn & Test"

    Directory of Open Access Journals (Sweden)

    Anca-Georgiana FODOR

    2016-11-01

    Full Text Available This article is aiming to present the architecture and few elements from the developing cycle of "English Grammar Learn & Test" app. This is an e-learning tool for people who want to improve their English Grammar and Vocabulary. The app was approved by Google Play and Apple Store and it is available for free on both platforms as following: Android: https://play.google.com/store/apps/details?id=com.labsterzz.english_tests iOS: https://itunes.apple.com/us/app/english-grammar-learn-test/id1126468980 The app already reached350.000 users, it is rated at 4.43out of maximum 5.0 in Google Play Store. Since mid-June 2016, we launched the app also in the Apple Store iOS devices.

  17. Voxel-wise prostate cell density prediction using multiparametric magnetic resonance imaging and machine learning.

    Science.gov (United States)

    Sun, Yu; Reynolds, Hayley M; Wraith, Darren; Williams, Scott; Finnegan, Mary E; Mitchell, Catherine; Murphy, Declan; Haworth, Annette

    2018-04-26

    There are currently no methods to estimate cell density in the prostate. This study aimed to develop predictive models to estimate prostate cell density from multiparametric magnetic resonance imaging (mpMRI) data at a voxel level using machine learning techniques. In vivo mpMRI data were collected from 30 patients before radical prostatectomy. Sequences included T2-weighted imaging, diffusion-weighted imaging and dynamic contrast-enhanced imaging. Ground truth cell density maps were computed from histology and co-registered with mpMRI. Feature extraction and selection were performed on mpMRI data. Final models were fitted using three regression algorithms including multivariate adaptive regression spline (MARS), polynomial regression (PR) and generalised additive model (GAM). Model parameters were optimised using leave-one-out cross-validation on the training data and model performance was evaluated on test data using root mean square error (RMSE) measurements. Predictive models to estimate voxel-wise prostate cell density were successfully trained and tested using the three algorithms. The best model (GAM) achieved a RMSE of 1.06 (± 0.06) × 10 3 cells/mm 2 and a relative deviation of 13.3 ± 0.8%. Prostate cell density can be quantitatively estimated non-invasively from mpMRI data using high-quality co-registered data at a voxel level. These cell density predictions could be used for tissue classification, treatment response evaluation and personalised radiotherapy.

  18. Mortality risk prediction in burn injury: Comparison of logistic regression with machine learning approaches.

    Science.gov (United States)

    Stylianou, Neophytos; Akbarov, Artur; Kontopantelis, Evangelos; Buchan, Iain; Dunn, Ken W

    2015-08-01

    Predicting mortality from burn injury has traditionally employed logistic regression models. Alternative machine learning methods have been introduced in some areas of clinical prediction as the necessary software and computational facilities have become accessible. Here we compare logistic regression and machine learning predictions of mortality from burn. An established logistic mortality model was compared to machine learning methods (artificial neural network, support vector machine, random forests and naïve Bayes) using a population-based (England & Wales) case-cohort registry. Predictive evaluation used: area under the receiver operating characteristic curve; sensitivity; specificity; positive predictive value and Youden's index. All methods had comparable discriminatory abilities, similar sensitivities, specificities and positive predictive values. Although some machine learning methods performed marginally better than logistic regression the differences were seldom statistically significant and clinically insubstantial. Random forests were marginally better for high positive predictive value and reasonable sensitivity. Neural networks yielded slightly better prediction overall. Logistic regression gives an optimal mix of performance and interpretability. The established logistic regression model of burn mortality performs well against more complex alternatives. Clinical prediction with a small set of strong, stable, independent predictors is unlikely to gain much from machine learning outside specialist research contexts. Copyright © 2015 Elsevier Ltd and ISBI. All rights reserved.

  19. The Future of Learning Technology: Some Tentative Predictions

    Science.gov (United States)

    Rushby, Nick

    2013-01-01

    This paper is a snapshot of an evolving vision of what the future may hold for learning technology. It offers three personal visions of the future and raises many questions that need to be explored if learning technology is to realise its full potential.

  20. Field tests and machine learning approaches for refining algorithms and correlations of driver's model parameters.

    Science.gov (United States)

    Tango, Fabio; Minin, Luca; Tesauri, Francesco; Montanari, Roberto

    2010-03-01

    This paper describes the field tests on a driving simulator carried out to validate the algorithms and the correlations of dynamic parameters, specifically driving task demand and drivers' distraction, able to predict drivers' intentions. These parameters belong to the driver's model developed by AIDE (Adaptive Integrated Driver-vehicle InterfacE) European Integrated Project. Drivers' behavioural data have been collected from the simulator tests to model and validate these parameters using machine learning techniques, specifically the adaptive neuro fuzzy inference systems (ANFIS) and the artificial neural network (ANN). Two models of task demand and distraction have been developed, one for each adopted technique. The paper provides an overview of the driver's model, the description of the task demand and distraction modelling and the tests conducted for the validation of these parameters. A test comparing predicted and expected outcomes of the modelled parameters for each machine learning technique has been carried out: for distraction, in particular, promising results (low prediction errors) have been obtained by adopting an artificial neural network.

  1. Predictive codes of familiarity and context during the perceptual learning of facial identities

    Science.gov (United States)

    Apps, Matthew A. J.; Tsakiris, Manos

    2013-11-01

    Face recognition is a key component of successful social behaviour. However, the computational processes that underpin perceptual learning and recognition as faces transition from unfamiliar to familiar are poorly understood. In predictive coding, learning occurs through prediction errors that update stimulus familiarity, but recognition is a function of both stimulus and contextual familiarity. Here we show that behavioural responses on a two-option face recognition task can be predicted by the level of contextual and facial familiarity in a computational model derived from predictive-coding principles. Using fMRI, we show that activity in the superior temporal sulcus varies with the contextual familiarity in the model, whereas activity in the fusiform face area covaries with the prediction error parameter that updated facial familiarity. Our results characterize the key computations underpinning the perceptual learning of faces, highlighting that the functional properties of face-processing areas conform to the principles of predictive coding.

  2. Towards Intelligence and Flexibility of Learning and Knowledge Testing Environments

    Directory of Open Access Journals (Sweden)

    Nerijus AUKSTAKALNIS

    2006-02-01

    Full Text Available The proposed goal oriented knowledge acquisition and assessment are based on the flexible educational model and allows to implement an adaptive control of the enhanced learning process according to the requirements of student's knowledge level, his state of cognition and subject learning history. The enhanced learner knowledge model specifies how the cognition state of the user will be achieved step by step. The use case actions definition is a starting point of the specification, which depends on different levels of learning scenarios and user cognition sub goals. The use case actions specification is used as a basis to set the requirements for service software specification and attributes of learning objects respectively. The paper presents the enhanced architecture of the student self-evaluation and on-line assessment system TestTool. The system is explored as an assessment engine capable of supporting and improving the individualized intelligent goal oriented self-instructional and simulation based mode of learning, grounded on the GRID distributed service architecture.

  3. Orchiopexy for intra-abdominal testes: factors predicting success.

    Science.gov (United States)

    Stec, Andrew A; Tanaka, Stacy T; Adams, Mark C; Pope, John C; Thomas, John C; Brock, John W

    2009-10-01

    Intra-abdominal testes can be treated with several surgical procedures. We evaluated factors influencing the outcome of orchiopexy for intra-abdominal testis. We retrospectively reviewed 156 consecutive orchiopexies performed for intra-abdominal testis, defined as a nonpalpable testis on examination and located in the abdomen at surgery. All surgical approaches were included in the study. Primary outcome was the overall success rate and secondary outcomes were success based on surgical approach, age and a patent processus vaginalis. Success was considered a testis with normal texture and size compared to the contralateral testis at followup. Multivariate analysis was performed to determine factors predictive of success. The overall success rate of all orchiopexies was 79.5%. Median patient age at orchiopexy was 12 months and mean followup was 16 months. Of the patients 117 had a patent processus vaginalis at surgery. One-stage abdominal orchiopexy was performed in 92 testes with 89.1% success. Of these cases 32 were performed laparoscopically with 96.9% success. One-stage Fowler-Stephens orchiopexy was performed in 27 testes and 2-stage Fowler-Stephens orchiopexy was performed in 37 with success in 63.0% and 67.6%, respectively. Multivariate analysis revealed that 1-stage orchiopexy without vessel division had more successful outcomes than 1 and 2-stage Fowler-Stephens orchiopexy (OR 0.24, p = 0.007 and 0.29, p = 0.19, respectively). Neither age at surgery nor an open internal ring was significant (p = 0.49 and 0.12, respectively). The overall success of orchiopexy for intra-abdominal testis is 79.5%. While patient selection remains a critical factor, 1-stage orchiopexy without vessel division was significantly more successful and a laparoscopic approach was associated with the fewest failures for intra-abdominal testes.

  4. The Trail Making test: a study of its ability to predict falls in the acute neurological in-patient population.

    Science.gov (United States)

    Mateen, Bilal Akhter; Bussas, Matthias; Doogan, Catherine; Waller, Denise; Saverino, Alessia; Király, Franz J; Playford, E Diane

    2018-05-01

    To determine whether tests of cognitive function and patient-reported outcome measures of motor function can be used to create a machine learning-based predictive tool for falls. Prospective cohort study. Tertiary neurological and neurosurgical center. In all, 337 in-patients receiving neurosurgical, neurological, or neurorehabilitation-based care. Binary (Y/N) for falling during the in-patient episode, the Trail Making Test (a measure of attention and executive function) and the Walk-12 (a patient-reported measure of physical function). The principal outcome was a fall during the in-patient stay ( n = 54). The Trail test was identified as the best predictor of falls. Moreover, addition of other variables, did not improve the prediction (Wilcoxon signed-rank P Test data (Wilcoxon signed-rank P test of cognitive function, the Trail Making test.

  5. Sorting Test, Tower Test and BRIEF-SR do not predict school performance of healthy adolescents in preuniversity education

    Directory of Open Access Journals (Sweden)

    Annemarie eBoschloo

    2014-04-01

    Full Text Available Executive functions (EF such as self-monitoring, planning and organizing are known to develop through childhood and adolescence. They are of potential importance for learning and school performance. Earlier research into the relation between executive functions and school performance did not provide clear results possibly because confounding factors such as educational track, boy-girl differences and parental education were not taken into account. The present study therefore investigated the relation between executive function tests and school performance in a highly controlled sample of 173 healthy adolescents aged 12-18. Only students in the pre-university educational track were used and the performance of boys was compared to that of girls. Results showed that there was no relation between the report marks obtained and the performance on executive function tests, notably the Sorting Test and the Tower Test of the Delis-Kaplan Executive Functions System (D-KEFS. Likewise, no relation was found between the report marks and the scores on the Behavior Rating Inventory of Executive Function – Self-Report Version (BRIEF-SR after these were controlled for grade, sex, and level of parental education.The findings indicate that executive functioning as measured with widely used instruments such as the BRIEF-SR does not predict school performance of adolescents in preuniversity education any better than a student’s grade, sex, and level of parental education. ed

  6. Non-linear dynamical signal characterization for prediction of defibrillation success through machine learning

    Directory of Open Access Journals (Sweden)

    Shandilya Sharad

    2012-10-01

    Full Text Available Abstract Background Ventricular Fibrillation (VF is a common presenting dysrhythmia in the setting of cardiac arrest whose main treatment is defibrillation through direct current countershock to achieve return of spontaneous circulation. However, often defibrillation is unsuccessful and may even lead to the transition of VF to more nefarious rhythms such as asystole or pulseless electrical activity. Multiple methods have been proposed for predicting defibrillation success based on examination of the VF waveform. To date, however, no analytical technique has been widely accepted. We developed a unique approach of computational VF waveform analysis, with and without addition of the signal of end-tidal carbon dioxide (PetCO2, using advanced machine learning algorithms. We compare these results with those obtained using the Amplitude Spectral Area (AMSA technique. Methods A total of 90 pre-countershock ECG signals were analyzed form an accessible preshosptial cardiac arrest database. A unified predictive model, based on signal processing and machine learning, was developed with time-series and dual-tree complex wavelet transform features. Upon selection of correlated variables, a parametrically optimized support vector machine (SVM model was trained for predicting outcomes on the test sets. Training and testing was performed with nested 10-fold cross validation and 6–10 features for each test fold. Results The integrative model performs real-time, short-term (7.8 second analysis of the Electrocardiogram (ECG. For a total of 90 signals, 34 successful and 56 unsuccessful defibrillations were classified with an average Accuracy and Receiver Operator Characteristic (ROC Area Under the Curve (AUC of 82.2% and 85%, respectively. Incorporation of the end-tidal carbon dioxide signal boosted Accuracy and ROC AUC to 83.3% and 93.8%, respectively, for a smaller dataset containing 48 signals. VF analysis using AMSA resulted in accuracy and ROC AUC of 64

  7. Precision Radiology: Predicting longevity using feature engineering and deep learning methods in a radiomics framework.

    Science.gov (United States)

    Oakden-Rayner, Luke; Carneiro, Gustavo; Bessen, Taryn; Nascimento, Jacinto C; Bradley, Andrew P; Palmer, Lyle J

    2017-05-10

    Precision medicine approaches rely on obtaining precise knowledge of the true state of health of an individual patient, which results from a combination of their genetic risks and environmental exposures. This approach is currently limited by the lack of effective and efficient non-invasive medical tests to define the full range of phenotypic variation associated with individual health. Such knowledge is critical for improved early intervention, for better treatment decisions, and for ameliorating the steadily worsening epidemic of chronic disease. We present proof-of-concept experiments to demonstrate how routinely acquired cross-sectional CT imaging may be used to predict patient longevity as a proxy for overall individual health and disease status using computer image analysis techniques. Despite the limitations of a modest dataset and the use of off-the-shelf machine learning methods, our results are comparable to previous 'manual' clinical methods for longevity prediction. This work demonstrates that radiomics techniques can be used to extract biomarkers relevant to one of the most widely used outcomes in epidemiological and clinical research - mortality, and that deep learning with convolutional neural networks can be usefully applied to radiomics research. Computer image analysis applied to routinely collected medical images offers substantial potential to enhance precision medicine initiatives.

  8. The validation and assessment of machine learning: a game of prediction from high-dimensional data.

    Directory of Open Access Journals (Sweden)

    Tune H Pers

    Full Text Available In applied statistics, tools from machine learning are popular for analyzing complex and high-dimensional data. However, few theoretical results are available that could guide to the appropriate machine learning tool in a new application. Initial development of an overall strategy thus often implies that multiple methods are tested and compared on the same set of data. This is particularly difficult in situations that are prone to over-fitting where the number of subjects is low compared to the number of potential predictors. The article presents a game which provides some grounds for conducting a fair model comparison. Each player selects a modeling strategy for predicting individual response from potential predictors. A strictly proper scoring rule, bootstrap cross-validation, and a set of rules are used to make the results obtained with different strategies comparable. To illustrate the ideas, the game is applied to data from the Nugenob Study where the aim is to predict the fat oxidation capacity based on conventional factors and high-dimensional metabolomics data. Three players have chosen to use support vector machines, LASSO, and random forests, respectively.

  9. Machine learning for predicting soil classes in three semi-arid landscapes

    Science.gov (United States)

    Brungard, Colby W.; Boettinger, Janis L.; Duniway, Michael C.; Wills, Skye A.; Edwards, Thomas C.

    2015-01-01

    Mapping the spatial distribution of soil taxonomic classes is important for informing soil use and management decisions. Digital soil mapping (DSM) can quantitatively predict the spatial distribution of soil taxonomic classes. Key components of DSM are the method and the set of environmental covariates used to predict soil classes. Machine learning is a general term for a broad set of statistical modeling techniques. Many different machine learning models have been applied in the literature and there are different approaches for selecting covariates for DSM. However, there is little guidance as to which, if any, machine learning model and covariate set might be optimal for predicting soil classes across different landscapes. Our objective was to compare multiple machine learning models and covariate sets for predicting soil taxonomic classes at three geographically distinct areas in the semi-arid western United States of America (southern New Mexico, southwestern Utah, and northeastern Wyoming). All three areas were the focus of digital soil mapping studies. Sampling sites at each study area were selected using conditioned Latin hypercube sampling (cLHS). We compared models that had been used in other DSM studies, including clustering algorithms, discriminant analysis, multinomial logistic regression, neural networks, tree based methods, and support vector machine classifiers. Tested machine learning models were divided into three groups based on model complexity: simple, moderate, and complex. We also compared environmental covariates derived from digital elevation models and Landsat imagery that were divided into three different sets: 1) covariates selected a priori by soil scientists familiar with each area and used as input into cLHS, 2) the covariates in set 1 plus 113 additional covariates, and 3) covariates selected using recursive feature elimination. Overall, complex models were consistently more accurate than simple or moderately complex models. Random

  10. Road Testing Graduate Attributes and Course Learning Outcomes of an Environmental Science Degree via a Work-Integrated Learning Placement

    Science.gov (United States)

    Whelan, Michael

    2017-01-01

    Graduate attributes and course learning outcomes are an integral part of higher education in Australia. Testing the performance of graduates in the workplace with regard to graduate attributes and course learning outcomes is a not a common occurrence. This study has road tested the graduate attributes and course learning outcomes of a bachelor…

  11. Predictive tests to evaluate oxidative potential of engineered nanomaterials

    Science.gov (United States)

    Ghiazza, Mara; Carella, Emanuele; Oliaro-Bosso, Simonetta; Corazzari, Ingrid; Viola, Franca; Fenoglio, Ivana

    2013-04-01

    Oxidative stress constitutes one of the principal injury mechanisms through which particulate toxicants (asbestos, crystalline silica, hard metals) and engineered nanomaterials can induce adverse health effects. ROS may be generated indirectly by activated cells and/or directly at the surface of the material. The occurrence of these processes depends upon the type of material. Many authors have recently demonstrated that metal oxides and carbon-based nanoparticles may influence (increasing or decreasing) the generation of oxygen radicals in a cell environment. Metal oxide, such as iron oxides, crystalline silica, and titanium dioxide are able to generate free radicals via different mechanisms causing an imbalance within oxidant species. The increase of ROS species may lead to inflammatory responses and in some cases to the development of cancer. On the other hand carbon-based nanomaterials, such as fullerene, carbon nanotubes, carbon black as well as cerium dioxide are able to scavenge the free radicals generated acting as antioxidant. The high numbers of new-engineered nanomaterials, which are introduced in the market, are exponentially increasing. Therefore the definition of toxicological strategies is urgently needed. The development of acellular screening tests will make possible the reduction of the number of in vitro and in vivo tests to be performed. An integrated protocol that may be used to predict the oxidant/antioxidant potential of engineered nanoparticles will be here presented.

  12. Analysing earthquake slip models with the spatial prediction comparison test

    KAUST Repository

    Zhang, L.; Mai, Paul Martin; Thingbaijam, Kiran Kumar; Razafindrakoto, H. N. T.; Genton, Marc G.

    2014-01-01

    Earthquake rupture models inferred from inversions of geophysical and/or geodetic data exhibit remarkable variability due to uncertainties in modelling assumptions, the use of different inversion algorithms, or variations in data selection and data processing. A robust statistical comparison of different rupture models obtained for a single earthquake is needed to quantify the intra-event variability, both for benchmark exercises and for real earthquakes. The same approach may be useful to characterize (dis-)similarities in events that are typically grouped into a common class of events (e.g. moderate-size crustal strike-slip earthquakes or tsunamigenic large subduction earthquakes). For this purpose, we examine the performance of the spatial prediction comparison test (SPCT), a statistical test developed to compare spatial (random) fields by means of a chosen loss function that describes an error relation between a 2-D field (‘model’) and a reference model. We implement and calibrate the SPCT approach for a suite of synthetic 2-D slip distributions, generated as spatial random fields with various characteristics, and then apply the method to results of a benchmark inversion exercise with known solution. We find the SPCT to be sensitive to different spatial correlations lengths, and different heterogeneity levels of the slip distributions. The SPCT approach proves to be a simple and effective tool for ranking the slip models with respect to a reference model.

  13. Analysing earthquake slip models with the spatial prediction comparison test

    KAUST Repository

    Zhang, L.

    2014-11-10

    Earthquake rupture models inferred from inversions of geophysical and/or geodetic data exhibit remarkable variability due to uncertainties in modelling assumptions, the use of different inversion algorithms, or variations in data selection and data processing. A robust statistical comparison of different rupture models obtained for a single earthquake is needed to quantify the intra-event variability, both for benchmark exercises and for real earthquakes. The same approach may be useful to characterize (dis-)similarities in events that are typically grouped into a common class of events (e.g. moderate-size crustal strike-slip earthquakes or tsunamigenic large subduction earthquakes). For this purpose, we examine the performance of the spatial prediction comparison test (SPCT), a statistical test developed to compare spatial (random) fields by means of a chosen loss function that describes an error relation between a 2-D field (‘model’) and a reference model. We implement and calibrate the SPCT approach for a suite of synthetic 2-D slip distributions, generated as spatial random fields with various characteristics, and then apply the method to results of a benchmark inversion exercise with known solution. We find the SPCT to be sensitive to different spatial correlations lengths, and different heterogeneity levels of the slip distributions. The SPCT approach proves to be a simple and effective tool for ranking the slip models with respect to a reference model.

  14. Predictive tests to evaluate oxidative potential of engineered nanomaterials

    International Nuclear Information System (INIS)

    Ghiazza, Mara; Carella, Emanuele; Corazzari, Ingrid; Fenoglio, Ivana; Oliaro-Bosso, Simonetta; Viola, Franca

    2013-01-01

    Oxidative stress constitutes one of the principal injury mechanisms through which particulate toxicants (asbestos, crystalline silica, hard metals) and engineered nanomaterials can induce adverse health effects. ROS may be generated indirectly by activated cells and/or directly at the surface of the material. The occurrence of these processes depends upon the type of material. Many authors have recently demonstrated that metal oxides and carbon-based nanoparticles may influence (increasing or decreasing) the generation of oxygen radicals in a cell environment. Metal oxide, such as iron oxides, crystalline silica, and titanium dioxide are able to generate free radicals via different mechanisms causing an imbalance within oxidant species. The increase of ROS species may lead to inflammatory responses and in some cases to the development of cancer. On the other hand carbon-based nanomaterials, such as fullerene, carbon nanotubes, carbon black as well as cerium dioxide are able to scavenge the free radicals generated acting as antioxidant. The high numbers of new-engineered nanomaterials, which are introduced in the market, are exponentially increasing. Therefore the definition of toxicological strategies is urgently needed. The development of acellular screening tests will make possible the reduction of the number of in vitro and in vivo tests to be performed. An integrated protocol that may be used to predict the oxidant/antioxidant potential of engineered nanoparticles will be here presented.

  15. Improving students' meaningful learning on the predictive nature of quantum mechanics

    Directory of Open Access Journals (Sweden)

    Rodolfo Alves de Carvalho Neto

    2009-03-01

    Full Text Available This paper deals with research about teaching quantum mechanics to 3rd year high school students and their meaningful learning of its predictive aspect; it is based on the Master’s dissertation of one of the authors (CARVALHO NETO, 2006. While teaching quantum mechanics, we emphasized its predictive and essentially probabilistic nature, based on Niels Bohr’s complementarity interpretation (BOHR, 1958. In this context, we have discussed the possibility of predicting measurement results in well-defined experimental contexts, even for individual events. Interviews with students reveal that they have used quantum mechanical ideas, suggesting their meaningful learning of the essentially probabilistic predictions of quantum mechanics.

  16. EEG Beta Power but Not Background Music Predicts the Recall Scores in a Foreign-Vocabulary Learning Task.

    Science.gov (United States)

    Küssner, Mats B; de Groot, Annette M B; Hofman, Winni F; Hillen, Marij A

    2016-01-01

    As tantalizing as the idea that background music beneficially affects foreign vocabulary learning may seem, there is-partly due to a lack of theory-driven research-no consistent evidence to support this notion. We investigated inter-individual differences in the effects of background music on foreign vocabulary learning. Based on Eysenck's theory of personality we predicted that individuals with a high level of cortical arousal should perform worse when learning with background music compared to silence, whereas individuals with a low level of cortical arousal should be unaffected by background music or benefit from it. Participants were tested in a paired-associate learning paradigm consisting of three immediate word recall tasks, as well as a delayed recall task one week later. Baseline cortical arousal assessed with spontaneous EEG measurement in silence prior to the learning rounds was used for the analyses. Results revealed no interaction between cortical arousal and the learning condition (background music vs. silence). Instead, we found an unexpected main effect of cortical arousal in the beta band on recall, indicating that individuals with high beta power learned more vocabulary than those with low beta power. To substantiate this finding we conducted an exact replication of the experiment. Whereas the main effect of cortical arousal was only present in a subsample of participants, a beneficial main effect of background music appeared. A combined analysis of both experiments suggests that beta power predicts the performance in the word recall task, but that there is no effect of background music on foreign vocabulary learning. In light of these findings, we discuss whether searching for effects of background music on foreign vocabulary learning, independent of factors such as inter-individual differences and task complexity, might be a red herring. Importantly, our findings emphasize the need for sufficiently powered research designs and exact replications

  17. EEG Beta Power but Not Background Music Predicts the Recall Scores in a Foreign-Vocabulary Learning Task.

    Directory of Open Access Journals (Sweden)

    Mats B Küssner

    Full Text Available As tantalizing as the idea that background music beneficially affects foreign vocabulary learning may seem, there is-partly due to a lack of theory-driven research-no consistent evidence to support this notion. We investigated inter-individual differences in the effects of background music on foreign vocabulary learning. Based on Eysenck's theory of personality we predicted that individuals with a high level of cortical arousal should perform worse when learning with background music compared to silence, whereas individuals with a low level of cortical arousal should be unaffected by background music or benefit from it. Participants were tested in a paired-associate learning paradigm consisting of three immediate word recall tasks, as well as a delayed recall task one week later. Baseline cortical arousal assessed with spontaneous EEG measurement in silence prior to the learning rounds was used for the analyses. Results revealed no interaction between cortical arousal and the learning condition (background music vs. silence. Instead, we found an unexpected main effect of cortical arousal in the beta band on recall, indicating that individuals with high beta power learned more vocabulary than those with low beta power. To substantiate this finding we conducted an exact replication of the experiment. Whereas the main effect of cortical arousal was only present in a subsample of participants, a beneficial main effect of background music appeared. A combined analysis of both experiments suggests that beta power predicts the performance in the word recall task, but that there is no effect of background music on foreign vocabulary learning. In light of these findings, we discuss whether searching for effects of background music on foreign vocabulary learning, independent of factors such as inter-individual differences and task complexity, might be a red herring. Importantly, our findings emphasize the need for sufficiently powered research designs and

  18. A comparison of machine learning techniques for predicting downstream acid mine drainage

    CSIR Research Space (South Africa)

    van Zyl, TL

    2014-07-01

    Full Text Available windowing approach over historical values to generate a prediction for the current value. We evaluate a number of Machine Learning techniques as regressors including Support Vector Regression, Random Forests, Stochastic Gradient Decent Regression, Linear...

  19. Applying machine learning to predict patient-specific current CD4 ...

    African Journals Online (AJOL)

    Apple apple

    This work shows the application of machine learning to predict current CD4 cell count of an HIV- .... Pre-processing ... remaining data elements of the PR and RT datasets. ... technique based on the structure of the human brain's neuron.

  20. Biological lifestyle factors in adult distance education: predicting cognitive and learning performance

    NARCIS (Netherlands)

    Gijselaers, Jérôme

    2015-01-01

    Gijselaers, H. J. M. (2015, 20 October). Biological lifestyle factors in adult distance education: predicting cognitive and learning performance. Presentation given for the inter-faculty Data Science group at the Open University of the Netherlands, Heerlen, The Netherlands.

  1. Multi-fidelity machine learning models for accurate bandgap predictions of solids

    International Nuclear Information System (INIS)

    Pilania, Ghanshyam; Gubernatis, James E.; Lookman, Turab

    2016-01-01

    Here, we present a multi-fidelity co-kriging statistical learning framework that combines variable-fidelity quantum mechanical calculations of bandgaps to generate a machine-learned model that enables low-cost accurate predictions of the bandgaps at the highest fidelity level. Additionally, the adopted Gaussian process regression formulation allows us to predict the underlying uncertainties as a measure of our confidence in the predictions. In using a set of 600 elpasolite compounds as an example dataset and using semi-local and hybrid exchange correlation functionals within density functional theory as two levels of fidelities, we demonstrate the excellent learning performance of the method against actual high fidelity quantum mechanical calculations of the bandgaps. The presented statistical learning method is not restricted to bandgaps or electronic structure methods and extends the utility of high throughput property predictions in a significant way.

  2. Smell test predicts performance on delayed recall memory test in elderly with depression.

    Science.gov (United States)

    Scalco, Monica Z; Streiner, David L; Rewilak, Dmytro; Castel, Saulo; Van Reekum, Robert

    2009-04-01

    Elderly with depression are at increased risk for cognitive dysfunction and dementia. Smell tests are correlated with performance on cognitive tests in the elderly and therefore might serve as a screening test for cognitive impairment in depressed elderly. To assess the validity of the CC-SIT (Cross-Cultural Smell Identification Test) as a screening test for cognitive impairment in elderly with depression. Forty-one patients, aged 60 and over, were assessed with the CC-SIT and CVLT (California Verbal Learning Test) after 3 months treatment of a Major Depressive Episode (DSM-IV) at the Day Hospital for Depression, Baycrest. Patients already diagnosed with dementia, or other psychiatric and neurological disorders, were excluded. Receiver Operating Characteristics (ROC) analysis was applied to assess the CC-SIT's accuracy in identifying individuals with impairment (2 SD below the mean for age and education or less) on CVLT delayed recall trials. Forty-one patients (33 women and eight men) were assessed. Mean age was 76.8 (SD: 6.5), mean HRSD scores before treatment was 22.0 (SD: 5.1). Nine patients had impairment on CVLT delayed recall measures. The area under the ROC curve was 0.776 (95% CI = 0.617-0.936). Our results support the use of the CC-SIT as a screening tool for cognitive impairment among elderly with depression as an indicator for the need of a comprehensive neuropsychological evaluation. Replication with larger samples is necessary. (c) 2008 John Wiley & Sons, Ltd.

  3. A Machine Learned Classifier That Uses Gene Expression Data to Accurately Predict Estrogen Receptor Status

    Science.gov (United States)

    Bastani, Meysam; Vos, Larissa; Asgarian, Nasimeh; Deschenes, Jean; Graham, Kathryn; Mackey, John; Greiner, Russell

    2013-01-01

    Background Selecting the appropriate treatment for breast cancer requires accurately determining the estrogen receptor (ER) status of the tumor. However, the standard for determining this status, immunohistochemical analysis of formalin-fixed paraffin embedded samples, suffers from numerous technical and reproducibility issues. Assessment of ER-status based on RNA expression can provide more objective, quantitative and reproducible test results. Methods To learn a parsimonious RNA-based classifier of hormone receptor status, we applied a machine learning tool to a training dataset of gene expression microarray data obtained from 176 frozen breast tumors, whose ER-status was determined by applying ASCO-CAP guidelines to standardized immunohistochemical testing of formalin fixed tumor. Results This produced a three-gene classifier that can predict the ER-status of a novel tumor, with a cross-validation accuracy of 93.17±2.44%. When applied to an independent validation set and to four other public databases, some on different platforms, this classifier obtained over 90% accuracy in each. In addition, we found that this prediction rule separated the patients' recurrence-free survival curves with a hazard ratio lower than the one based on the IHC analysis of ER-status. Conclusions Our efficient and parsimonious classifier lends itself to high throughput, highly accurate and low-cost RNA-based assessments of ER-status, suitable for routine high-throughput clinical use. This analytic method provides a proof-of-principle that may be applicable to developing effective RNA-based tests for other biomarkers and conditions. PMID:24312637

  4. A machine learned classifier that uses gene expression data to accurately predict estrogen receptor status.

    Directory of Open Access Journals (Sweden)

    Meysam Bastani

    Full Text Available BACKGROUND: Selecting the appropriate treatment for breast cancer requires accurately determining the estrogen receptor (ER status of the tumor. However, the standard for determining this status, immunohistochemical analysis of formalin-fixed paraffin embedded samples, suffers from numerous technical and reproducibility issues. Assessment of ER-status based on RNA expression can provide more objective, quantitative and reproducible test results. METHODS: To learn a parsimonious RNA-based classifier of hormone receptor status, we applied a machine learning tool to a training dataset of gene expression microarray data obtained from 176 frozen breast tumors, whose ER-status was determined by applying ASCO-CAP guidelines to standardized immunohistochemical testing of formalin fixed tumor. RESULTS: This produced a three-gene classifier that can predict the ER-status of a novel tumor, with a cross-validation accuracy of 93.17±2.44%. When applied to an independent validation set and to four other public databases, some on different platforms, this classifier obtained over 90% accuracy in each. In addition, we found that this prediction rule separated the patients' recurrence-free survival curves with a hazard ratio lower than the one based on the IHC analysis of ER-status. CONCLUSIONS: Our efficient and parsimonious classifier lends itself to high throughput, highly accurate and low-cost RNA-based assessments of ER-status, suitable for routine high-throughput clinical use. This analytic method provides a proof-of-principle that may be applicable to developing effective RNA-based tests for other biomarkers and conditions.

  5. Predicting cyberbullying perpetration in emerging adults: A theoretical test of the Barlett Gentile Cyberbullying Model.

    Science.gov (United States)

    Barlett, Christopher; Chamberlin, Kristina; Witkower, Zachary

    2017-04-01

    The Barlett and Gentile Cyberbullying Model (BGCM) is a learning-based theory that posits the importance of positive cyberbullying attitudes predicting subsequent cyberbullying perpetration. Furthermore, the tenants of the BGCM state that cyberbullying attitude are likely to form when the online aggressor believes that the online environment allows individuals of all physical sizes to harm others and they are perceived as anonymous. Past work has tested parts of the BGCM; no study has used longitudinal methods to examine this model fully. The current study (N = 161) employed a three-wave longitudinal design to test the BGCM. Participants (age range: 18-24) completed measures of the belief that physical strength is irrelevant online and anonymity perceptions at Wave 1, cyberbullying attitudes at Wave 2, and cyberbullying perpetration at Wave 3. Results showed strong support for the BGCM: anonymity perceptions and the belief that physical attributes are irrelevant online at Wave 1 predicted Wave 2 cyberbullying attitudes, which predicted subsequent Wave 3 cyberbullying perpetration. These results support the BGCM and are the first to show empirical support for this model. Aggr. Behav. 43:147-154, 2017. © 2016 Wiley Periodicals, Inc. © 2016 Wiley Periodicals, Inc.

  6. Sex differences in cognitive ageing: testing predictions derived from life-history theory in a dioecious nematode.

    Science.gov (United States)

    Zwoinska, Martyna K; Kolm, Niclas; Maklakov, Alexei A

    2013-12-01

    Life-history theory maintains that organisms allocate limited resources to different traits to maximize fitness. Learning ability and memory are costly and known to trade-off with longevity in invertebrates. However, since the relationship between longevity and fitness often differs between the sexes, it is likely that sexes will differentially resolve the trade-off between learning and longevity. We used an established associative learning paradigm in the dioecious nematode Caenorhabditis remanei, which is sexually dimorphic for lifespan, to study age-related learning ability in males and females. In particular, we tested the hypothesis that females (the shorter-lived sex) show higher learning ability than males early in life but senesce faster. Indeed, young females outperformed young males in learning a novel association between an odour (butanone) and food (bacteria). However, while learning ability and offspring production declined rapidly with age in females, males maintained high levels of these traits until mid-age. These results not only demonstrate sexual dimorphism in age-related learning ability but also suggest that it conforms to predictions derived from the life-history theory. © 2013.

  7. Predicting Bobsled Pushing Ability from Various Combine Testing Events.

    Science.gov (United States)

    Tomasevicz, Curtis L; Ransone, Jack W; Bach, Christopher W

    2018-03-12

    The requisite combination of speed, power, and strength necessary for a bobsled push athlete coupled with the difficulty in directly measuring pushing ability makes selecting effective push crews challenging. Current practices by USA Bobsled and Skeleton (USABS) utilize field combine testing to assess and identify specifically selected performance variables in an attempt to best predict push performance abilities. Combine data consisting of 11 physical performance variables were collected from 75 subjects across two winter Olympic qualification years (2009 and 2013). These variables were sprints of 15-, 30-, and 60 m, a flying 30 m sprint, a standing broad jump, a shot toss, squat, power clean, body mass, and dry-land brake and side bobsled pushes. Discriminant Analysis (DA) in addition to Principle Component Analysis (PCA) was used to investigate two cases (Case 1: Olympians vs. non-Olympians; Case 2: National Team vs. non-National Team). Using these 11 variables, DA led to a classification rule that proved capable of identifying Olympians from non-Olympians and National Team members from non-National Team members with 9.33% and 14.67% misclassification rates, respectively. The PCA was used to find similar test variables within the combine that provided redundant or useless data. After eliminating the unnecessary variables, DA on the new combinations showed that 8 (Case 1) and 20 (Case 2) other combinations with fewer performance variables yielded misclassification rates as low as 6.67% and 13.33% respectively. Utilizing fewer performance variables can allow governing bodies in many other sports to create more appropriate combine testing that maximize accuracy, while minimizing irrelevant and redundant strategies.

  8. Applying Learning Analytics for the Early Prediction of Students' Academic Performance in Blended Learning

    Science.gov (United States)

    Lu, Owen H. T.; Huang, Anna Y. Q.; Huang, Jeff C. H.; Lin, Albert J. Q.; Ogata, Hiroaki; Yang, Stephen J. H.

    2018-01-01

    Blended learning combines online digital resources with traditional classroom activities and enables students to attain higher learning performance through well-defined interactive strategies involving online and traditional learning activities. Learning analytics is a conceptual framework and is a part of our Precision education used to analyze…

  9. Individual differences in learning predict the return of fear.

    Science.gov (United States)

    Gershman, Samuel J; Hartley, Catherine A

    2015-09-01

    Using a laboratory analogue of learned fear (Pavlovian fear conditioning), we show that there is substantial heterogeneity across individuals in spontaneous recovery of fear following extinction training. We propose that this heterogeneity might stem from qualitative individual differences in the nature of extinction learning. Whereas some individuals tend to form a new memory during extinction, leaving their fear memory intact, others update the original threat association with new safety information, effectively unlearning the fear memory. We formalize this account in a computational model of fear learning and show that individuals who, according to the model, are more likely to form new extinction memories tend to show greater spontaneous recovery compared to individuals who appear to only update a single memory. This qualitative variation in fear and extinction learning may have important implications for understanding vulnerability and resilience to fear-related psychiatric disorders.

  10. Building Customer Churn Prediction Models in Fitness Industry with Machine Learning Methods

    OpenAIRE

    Shan, Min

    2017-01-01

    With the rapid growth of digital systems, churn management has become a major focus within customer relationship management in many industries. Ample research has been conducted for churn prediction in different industries with various machine learning methods. This thesis aims to combine feature selection and supervised machine learning methods for defining models of churn prediction and apply them on fitness industry. Forward selection is chosen as feature selection methods. Support Vector ...

  11. Hebbian learning and predictive mirror neurons for actions, sensations and emotions

    OpenAIRE

    Keysers, C.; Gazzola, Valeria

    2014-01-01

    Spike-timing-dependent plasticity is considered the neurophysiological basis of Hebbian learning and has been shown to be sensitive to both contingency and contiguity between pre- and postsynaptic activity. Here, we will examine how applying this Hebbian learning rule to a system of interconnected neurons in the presence of direct or indirect re-afference (e.g. seeing/hearing one's own actions) predicts the emergence of mirror neurons with predictive properties. In this framework, we analyse ...

  12. Learning Type Extension Trees for Metal Bonding State Prediction

    DEFF Research Database (Denmark)

    Frasconi, Paolo; Jaeger, Manfred; Passerini, Andrea

    2008-01-01

    Type Extension Trees (TET) have been recently introduced as an expressive representation language allowing to encode complex combinatorial features of relational entities. They can be efficiently learned with a greedy search strategy driven by a generalized relational information gain and a discr......Type Extension Trees (TET) have been recently introduced as an expressive representation language allowing to encode complex combinatorial features of relational entities. They can be efficiently learned with a greedy search strategy driven by a generalized relational information gain...

  13. Feature selection in wind speed prediction systems based on a hybrid coral reefs optimization – Extreme learning machine approach

    International Nuclear Information System (INIS)

    Salcedo-Sanz, S.; Pastor-Sánchez, A.; Prieto, L.; Blanco-Aguilera, A.; García-Herrera, R.

    2014-01-01

    Highlights: • A novel approach for short-term wind speed prediction is presented. • The system is formed by a coral reefs optimization algorithm and an extreme learning machine. • Feature selection is carried out with the CRO to improve the ELM performance. • The method is tested in real wind farm data in USA, for the period 2007–2008. - Abstract: This paper presents a novel approach for short-term wind speed prediction based on a Coral Reefs Optimization algorithm (CRO) and an Extreme Learning Machine (ELM), using meteorological predictive variables from a physical model (the Weather Research and Forecast model, WRF). The approach is based on a Feature Selection Problem (FSP) carried out with the CRO, that must obtain a reduced number of predictive variables out of the total available from the WRF. This set of features will be the input of an ELM, that finally provides the wind speed prediction. The CRO is a novel bio-inspired approach, based on the simulation of reef formation and coral reproduction, able to obtain excellent results in optimization problems. On the other hand, the ELM is a new paradigm in neural networks’ training, that provides a robust and extremely fast training of the network. Together, these algorithms are able to successfully solve this problem of feature selection in short-term wind speed prediction. Experiments in a real wind farm in the USA show the excellent performance of the CRO–ELM approach in this FSP wind speed prediction problem

  14. Beyond Engagement Analytics: Which Online Mixed-Data Factors Predict Student Learning Outcomes?

    Science.gov (United States)

    Strang, Kenneth David

    2017-01-01

    This mixed-method study focuses on online learning analytics, a research area of importance. Several important student attributes and their online activities are examined to identify what seems to work best to predict higher grades. The purpose is to explore the relationships between student grade and key learning engagement factors using a large…

  15. Translating visual information into action predictions: Statistical learning in action and nonaction contexts.

    Science.gov (United States)

    Monroy, Claire D; Gerson, Sarah A; Hunnius, Sabine

    2018-05-01

    Humans are sensitive to the statistical regularities in action sequences carried out by others. In the present eyetracking study, we investigated whether this sensitivity can support the prediction of upcoming actions when observing unfamiliar action sequences. In two between-subjects conditions, we examined whether observers would be more sensitive to statistical regularities in sequences performed by a human agent versus self-propelled 'ghost' events. Secondly, we investigated whether regularities are learned better when they are associated with contingent effects. Both implicit and explicit measures of learning were compared between agent and ghost conditions. Implicit learning was measured via predictive eye movements to upcoming actions or events, and explicit learning was measured via both uninstructed reproduction of the action sequences and verbal reports of the regularities. The findings revealed that participants, regardless of condition, readily learned the regularities and made correct predictive eye movements to upcoming events during online observation. However, different patterns of explicit-learning outcomes emerged following observation: Participants were most likely to re-create the sequence regularities and to verbally report them when they had observed an actor create a contingent effect. These results suggest that the shift from implicit predictions to explicit knowledge of what has been learned is facilitated when observers perceive another agent's actions and when these actions cause effects. These findings are discussed with respect to the potential role of the motor system in modulating how statistical regularities are learned and used to modify behavior.

  16. Participatory cues and program familiarity predict young children’s learning from educational television

    NARCIS (Netherlands)

    Piotrowski, J.

    2014-01-01

    The capacity model is designed to predict young children's learning from educational television. It posits that select program features and individual child characteristics can support this learning either by increasing total working memory allocated to the program or altering the allocation of

  17. Classroom Organization and Teacher Stress Predict Learning Motivation in Kindergarten Children

    Science.gov (United States)

    Pakarinen, Eija; Kiuru, Noona; Lerkkanen, Marja-Kristiina; Poikkeus, Anna-Maija; Siekkinen, Martti; Nurmi, Jari-Erik

    2010-01-01

    This study examined the extent to which observed teaching practices and self-reported teacher stress predict children's learning motivation and phonological awareness in kindergarten. The pre-reading skills of 1,268 children were measured at the beginning of their kindergarten year. Their learning motivation and phonological awareness were…

  18. Dynamic Models of Learning That Characterize Parent-Child Exchanges Predict Vocabulary Growth

    Science.gov (United States)

    Ober, David R.; Beekman, John A.

    2016-01-01

    Cumulative vocabulary models for infants and toddlers were developed from models of learning that predict trajectories associated with low, average, and high vocabulary growth rates (14 to 46 months). It was hypothesized that models derived from rates of learning mirror the type of exchanges provided to infants and toddlers by parents and…

  19. Native-language N400 and P600 predict dissociable language-learning abilities in adults.

    Science.gov (United States)

    Qi, Zhenghan; Beach, Sara D; Finn, Amy S; Minas, Jennifer; Goetz, Calvin; Chan, Brian; Gabrieli, John D E

    2017-04-01

    Language learning aptitude during adulthood varies markedly across individuals. An individual's native-language ability has been associated with success in learning a new language as an adult. However, little is known about how native-language processing affects learning success and what neural markers of native-language processing, if any, are related to success in learning. We therefore related variation in electrophysiology during native-language processing to success in learning a novel artificial language. Event-related potentials (ERPs) were recorded while native English speakers judged the acceptability of English sentences prior to learning an artificial language. There was a trend towards a double dissociation between native-language ERPs and their relationships to novel syntax and vocabulary learning. Individuals who exhibited a greater N400 effect when processing English semantics showed better future learning of the artificial language overall. The N400 effect was related to syntax learning via its specific relationship to vocabulary learning. In contrast, the P600 effect size when processing English syntax predicted future syntax learning but not vocabulary learning. These findings show that distinct neural signatures of native-language processing relate to dissociable abilities for learning novel semantic and syntactic information. Copyright © 2016 Elsevier Ltd. All rights reserved.

  20. Native-language N400 and P600 predict dissociable language-learning abilities in adults

    Science.gov (United States)

    Qi, Zhenghan; Beach, Sara D.; Finn, Amy S.; Minas, Jennifer; Goetz, Calvin; Chan, Brian; Gabrieli, John D.E.

    2018-01-01

    Language learning aptitude during adulthood varies markedly across individuals. An individual’s native-language ability has been associated with success in learning a new language as an adult. However, little is known about how native-language processing affects learning success and what neural markers of native-language processing, if any, are related to success in learning. We therefore related variation in electrophysiology during native-language processing to success in learning a novel artificial language. Event-related potentials (ERPs) were recorded while native English speakers judged the acceptability of English sentences prior to learning an artificial language. There was a trend towards a double dissociation between native-language ERPs and their relationships to novel syntax and vocabulary learning. Individuals who exhibited a greater N400 effect when processing English semantics showed better future learning of the artificial language overall. The N400 effect was related to syntax learning via its specific relationship to vocabulary learning. In contrast, the P600 effect size when processing English syntax predicted future syntax learning but not vocabulary learning. These findings show that distinct neural signatures of native-language processing relate to dissociable abilities for learning novel semantic and syntactic information. PMID:27737775

  1. Implementation and testing of the travel time prediction system (TIPS)

    OpenAIRE

    PANT, Prahlad D; UNIVERSITY OF CINCINNATI

    2001-01-01

    RAPPORT DE RECHERCHE FINAL The Travel Time Prediction System (TIPS) is a portable automated system for predicting and displaying travel time for motorists in advance of and through freeway construction work zones,on a real-time basis

  2. Roles of dopamine neurons in mediating the prediction error in aversive learning in insects.

    Science.gov (United States)

    Terao, Kanta; Mizunami, Makoto

    2017-10-31

    In associative learning in mammals, it is widely accepted that the discrepancy, or error, between actual and predicted reward determines whether learning occurs. The prediction error theory has been proposed to account for the finding of a blocking phenomenon, in which pairing of a stimulus X with an unconditioned stimulus (US) could block subsequent association of a second stimulus Y to the US when the two stimuli were paired in compound with the same US. Evidence for this theory, however, has been imperfect since blocking can also be accounted for by competitive theories. We recently reported blocking in classical conditioning of an odor with water reward in crickets. We also reported an "auto-blocking" phenomenon in appetitive learning, which supported the prediction error theory and rejected alternative theories. The presence of auto-blocking also suggested that octopamine neurons mediate reward prediction error signals. Here we show that blocking and auto-blocking occur in aversive learning to associate an odor with salt water (US) in crickets, and our results suggest that dopamine neurons mediate aversive prediction error signals. We conclude that the prediction error theory is applicable to both appetitive learning and aversive learning in insects.

  3. Prognostic durability of liver fibrosis tests and improvement in predictive performance for mortality by combining tests.

    Science.gov (United States)

    Bertrais, Sandrine; Boursier, Jérôme; Ducancelle, Alexandra; Oberti, Frédéric; Fouchard-Hubert, Isabelle; Moal, Valérie; Calès, Paul

    2017-06-01

    There is currently no recommended time interval between noninvasive fibrosis measurements for monitoring chronic liver diseases. We determined how long a single liver fibrosis evaluation may accurately predict mortality, and assessed whether combining tests improves prognostic performance. We included 1559 patients with chronic liver disease and available baseline liver stiffness measurement (LSM) by Fibroscan, aspartate aminotransferase to platelet ratio index (APRI), FIB-4, Hepascore, and FibroMeter V2G . Median follow-up was 2.8 years during which 262 (16.8%) patients died, with 115 liver-related deaths. All fibrosis tests were able to predict mortality, although APRI (and FIB-4 for liver-related mortality) showed lower overall discriminative ability than the other tests (differences in Harrell's C-index: P fibrosis, 1 year in patients with significant fibrosis, and liver disease (MELD) score testing sets. In the training set, blood tests and LSM were independent predictors of all-cause mortality. The best-fit multivariate model included age, sex, LSM, and FibroMeter V2G with C-index = 0.834 (95% confidence interval, 0.803-0.862). The prognostic model for liver-related mortality included the same covariates with C-index = 0.868 (0.831-0.902). In the testing set, the multivariate models had higher prognostic accuracy than FibroMeter V2G or LSM alone for all-cause mortality and FibroMeter V2G alone for liver-related mortality. The prognostic durability of a single baseline fibrosis evaluation depends on the liver fibrosis level. Combining LSM with a blood fibrosis test improves mortality risk assessment. © 2016 Journal of Gastroenterology and Hepatology Foundation and John Wiley & Sons Australia, Ltd.

  4. Application of Machine Learning Approaches for Protein-protein Interactions Prediction.

    Science.gov (United States)

    Zhang, Mengying; Su, Qiang; Lu, Yi; Zhao, Manman; Niu, Bing

    2017-01-01

    Proteomics endeavors to study the structures, functions and interactions of proteins. Information of the protein-protein interactions (PPIs) helps to improve our knowledge of the functions and the 3D structures of proteins. Thus determining the PPIs is essential for the study of the proteomics. In this review, in order to study the application of machine learning in predicting PPI, some machine learning approaches such as support vector machine (SVM), artificial neural networks (ANNs) and random forest (RF) were selected, and the examples of its applications in PPIs were listed. SVM and RF are two commonly used methods. Nowadays, more researchers predict PPIs by combining more than two methods. This review presents the application of machine learning approaches in predicting PPI. Many examples of success in identification and prediction in the area of PPI prediction have been discussed, and the PPIs research is still in progress. Copyright© Bentham Science Publishers; For any queries, please email at epub@benthamscience.org.

  5. Life Prediction for FRP composites with Data Fusion & Machine Learning

    Data.gov (United States)

    National Aeronautics and Space Administration — High-fidelity, probabilistic predictions of damage evolution in fiber-reinforced polymer (FRP) composite structures could accelerate development and certification of...

  6. The impact of predictive genetic testing for hereditary nonpolyposis colorectal cancer: three years after testing.

    Science.gov (United States)

    Collins, Veronica R; Meiser, Bettina; Ukoumunne, Obioha C; Gaff, Clara; St John, D James; Halliday, Jane L

    2007-05-01

    To fully assess predictive genetic testing programs, it is important to assess outcomes over periods of time longer than the 1-year follow-up reported in the literature. We conducted a 3-year study of individuals who received predictive genetic test results for previously identified familial mutations in Australian Familial Cancer Clinics. Questionnaires were sent before attendance at the familial cancer clinic and 2 weeks, 4 months, 1 year, and 3 years after receiving test results. Psychological measures were included each time, and preventive behaviors were assessed at baseline and 1 and 3 years. Psychological measures were adjusted for age, gender, and baseline score. The study included 19 carriers and 54 non-carriers. We previously reported an increase in mean cancer-specific distress in carriers at 2 weeks with a return to baseline levels by 12 months. This level was maintained until 3 years. Non-carriers showed sustained decreases after testing with a significantly lower level at 3 years compared with baseline (P depression and anxiety scores did not differ between carriers and non-carriers and, at 3 years, were similar to baseline. All carriers and 7% of non-carriers had had a colonoscopy by 3 years, and 69% of 13 female carriers had undergone gynecological screening in the previous 2 years. Prophylactic surgery was rare. This report of long-term data indicates appropriate screening and improved psychological measures for non-carriers with no evidence of undue psychological distress in carriers of hereditary nonpolyposis colorectal cancer mutations.

  7. Value of supervised learning events in predicting doctors in difficulty.

    Science.gov (United States)

    Patel, Mumtaz; Agius, Steven; Wilkinson, Jack; Patel, Leena; Baker, Paul

    2016-07-01

    In the UK, supervised learning events (SLE) replaced traditional workplace-based assessments for foundation-year trainees in 2012. A key element of SLEs was to incorporate trainee reflection and assessor feedback in order to drive learning and identify training issues early. Few studies, however, have investigated the value of SLEs in predicting doctors in difficulty. This study aimed to identify principles that would inform understanding about how and why SLEs work or not in identifying doctors in difficulty (DiD). A retrospective case-control study of North West Foundation School trainees' electronic portfolios was conducted. Cases comprised all known DiD. Controls were randomly selected from the same cohort. Free-text supervisor comments from each SLE were assessed for the four domains defined in the General Medical Council's Good Medical Practice Guidelines and each scored blindly for level of concern using a three-point ordinal scale. Cumulative scores for each SLE were then analysed quantitatively for their predictive value of actual DiD. A qualitative thematic analysis was also conducted. The prevalence of DiD in this sample was 6.5%. Receiver operator characteristic curve analysis showed that Team Assessment of Behaviour (TAB) was the only SLE strongly predictive of actual DiD status. The Educational Supervisor Report (ESR) was also strongly predictive of DiD status. Fisher's test showed significant associations of TAB and ESR for both predicted and actual DiD status and also the health and performance subtypes. None of the other SLEs showed significant associations. Qualitative data analysis revealed inadequate completion and lack of constructive, particularly negative, feedback. This indicated that SLEs were not used to their full potential. TAB and the ESR are strongly predictive of DiD. However, SLEs are not being used to their full potential, and the quality of completion of reports on SLEs and feedback needs to be improved in order to better identify

  8. Prediction of revascularization after myocardial perfusion SPECT by machine learning in a large population.

    Science.gov (United States)

    Arsanjani, Reza; Dey, Damini; Khachatryan, Tigran; Shalev, Aryeh; Hayes, Sean W; Fish, Mathews; Nakanishi, Rine; Germano, Guido; Berman, Daniel S; Slomka, Piotr

    2015-10-01

    We aimed to investigate if early revascularization in patients with suspected coronary artery disease can be effectively predicted by integrating clinical data and quantitative image features derived from perfusion SPECT (MPS) by machine learning (ML) approach. 713 rest (201)Thallium/stress (99m)Technetium MPS studies with correlating invasive angiography with 372 revascularization events (275 PCI/97 CABG) within 90 days after MPS (91% within 30 days) were considered. Transient ischemic dilation, stress combined supine/prone total perfusion deficit (TPD), supine rest and stress TPD, exercise ejection fraction, and end-systolic volume, along with clinical parameters including patient gender, history of hypertension and diabetes mellitus, ST-depression on baseline ECG, ECG and clinical response during stress, and post-ECG probability by boosted ensemble ML algorithm (LogitBoost) to predict revascularization events. These features were selected using an automated feature selection algorithm from all available clinical and quantitative data (33 parameters). Tenfold cross-validation was utilized to train and test the prediction model. The prediction of revascularization by ML algorithm was compared to standalone measures of perfusion and visual analysis by two experienced readers utilizing all imaging, quantitative, and clinical data. The sensitivity of machine learning (ML) (73.6% ± 4.3%) for prediction of revascularization was similar to one reader (73.9% ± 4.6%) and standalone measures of perfusion (75.5% ± 4.5%). The specificity of ML (74.7% ± 4.2%) was also better than both expert readers (67.2% ± 4.9% and 66.0% ± 5.0%, P < .05), but was similar to ischemic TPD (68.3% ± 4.9%, P < .05). The receiver operator characteristics areas under curve for ML (0.81 ± 0.02) was similar to reader 1 (0.81 ± 0.02) but superior to reader 2 (0.72 ± 0.02, P < .01) and standalone measure of perfusion (0.77 ± 0.02, P < .01). ML approach is comparable or better than

  9. Dispositional optimism and perceived risk interact to predict intentions to learn genome sequencing results.

    Science.gov (United States)

    Taber, Jennifer M; Klein, William M P; Ferrer, Rebecca A; Lewis, Katie L; Biesecker, Leslie G; Biesecker, Barbara B

    2015-07-01

    Dispositional optimism and risk perceptions are each associated with health-related behaviors and decisions and other outcomes, but little research has examined how these constructs interact, particularly in consequential health contexts. The predictive validity of risk perceptions for health-related information seeking and intentions may be improved by examining dispositional optimism as a moderator, and by testing alternate types of risk perceptions, such as comparative and experiential risk. Participants (n = 496) had their genomes sequenced as part of a National Institutes of Health pilot cohort study (ClinSeq®). Participants completed a cross-sectional baseline survey of various types of risk perceptions and intentions to learn genome sequencing results for differing disease risks (e.g., medically actionable, nonmedically actionable, carrier status) and to use this information to change their lifestyle/health behaviors. Risk perceptions (absolute, comparative, and experiential) were largely unassociated with intentions to learn sequencing results. Dispositional optimism and comparative risk perceptions interacted, however, such that individuals higher in optimism reported greater intentions to learn all 3 types of sequencing results when comparative risk was perceived to be higher than when it was perceived to be lower. This interaction was inconsistent for experiential risk and absent for absolute risk. Independent of perceived risk, participants high in dispositional optimism reported greater interest in learning risks for nonmedically actionable disease and carrier status, and greater intentions to use genome information to change their lifestyle/health behaviors. The relationship between risk perceptions and intentions may depend on how risk perceptions are assessed and on degree of optimism. (c) 2015 APA, all rights reserved.

  10. Osteoporosis risk prediction for bone mineral density assessment of postmenopausal women using machine learning.

    Science.gov (United States)

    Yoo, Tae Keun; Kim, Sung Kean; Kim, Deok Won; Choi, Joon Yul; Lee, Wan Hyung; Oh, Ein; Park, Eun-Cheol

    2013-11-01

    A number of clinical decision tools for osteoporosis risk assessment have been developed to select postmenopausal women for the measurement of bone mineral density. We developed and validated machine learning models with the aim of more accurately identifying the risk of osteoporosis in postmenopausal women compared to the ability of conventional clinical decision tools. We collected medical records from Korean postmenopausal women based on the Korea National Health and Nutrition Examination Surveys. The training data set was used to construct models based on popular machine learning algorithms such as support vector machines (SVM), random forests, artificial neural networks (ANN), and logistic regression (LR) based on simple surveys. The machine learning models were compared to four conventional clinical decision tools: osteoporosis self-assessment tool (OST), osteoporosis risk assessment instrument (ORAI), simple calculated osteoporosis risk estimation (SCORE), and osteoporosis index of risk (OSIRIS). SVM had significantly better area under the curve (AUC) of the receiver operating characteristic than ANN, LR, OST, ORAI, SCORE, and OSIRIS for the training set. SVM predicted osteoporosis risk with an AUC of 0.827, accuracy of 76.7%, sensitivity of 77.8%, and specificity of 76.0% at total hip, femoral neck, or lumbar spine for the testing set. The significant factors selected by SVM were age, height, weight, body mass index, duration of menopause, duration of breast feeding, estrogen therapy, hyperlipidemia, hypertension, osteoarthritis, and diabetes mellitus. Considering various predictors associated with low bone density, the machine learning methods may be effective tools for identifying postmenopausal women at high risk for osteoporosis.

  11. Individual personality differences in goats predict their performance in visual learning and non-associative cognitive tasks.

    Science.gov (United States)

    Nawroth, Christian; Prentice, Pamela M; McElligott, Alan G

    2017-01-01

    Variation in common personality traits, such as boldness or exploration, is often associated with risk-reward trade-offs and behavioural flexibility. To date, only a few studies have examined the effects of consistent behavioural traits on both learning and cognition. We investigated whether certain personality traits ('exploration' and 'sociability') of individuals were related to cognitive performance, learning flexibility and learning style in a social ungulate species, the goat (Capra hircus). We also investigated whether a preference for feature cues rather than impaired learning abilities can explain performance variation in a visual discrimination task. We found that personality scores were consistent across time and context. Less explorative goats performed better in a non-associative cognitive task, in which subjects had to follow the trajectory of a hidden object (i.e. testing their ability for object permanence). We also found that less sociable subjects performed better compared to more sociable goats in a visual discrimination task. Good visual learning performance was associated with a preference for feature cues, indicating personality-dependent learning strategies in goats. Our results suggest that personality traits predict the outcome in visual discrimination and non-associative cognitive tasks in goats and that impaired performance in a visual discrimination tasks does not necessarily imply impaired learning capacities, but rather can be explained by a varying preference for feature cues. Copyright © 2016 Elsevier B.V. All rights reserved.

  12. Prediction of Bispectral Index during Target-controlled Infusion of Propofol and Remifentanil: A Deep Learning Approach.

    Science.gov (United States)

    Lee, Hyung-Chul; Ryu, Ho-Geol; Chung, Eun-Jin; Jung, Chul-Woo

    2018-03-01

    The discrepancy between predicted effect-site concentration and measured bispectral index is problematic during intravenous anesthesia with target-controlled infusion of propofol and remifentanil. We hypothesized that bispectral index during total intravenous anesthesia would be more accurately predicted by a deep learning approach. Long short-term memory and the feed-forward neural network were sequenced to simulate the pharmacokinetic and pharmacodynamic parts of an empirical model, respectively, to predict intraoperative bispectral index during combined use of propofol and remifentanil. Inputs of long short-term memory were infusion histories of propofol and remifentanil, which were retrieved from target-controlled infusion pumps for 1,800 s at 10-s intervals. Inputs of the feed-forward network were the outputs of long short-term memory and demographic data such as age, sex, weight, and height. The final output of the feed-forward network was the bispectral index. The performance of bispectral index prediction was compared between the deep learning model and previously reported response surface model. The model hyperparameters comprised 8 memory cells in the long short-term memory layer and 16 nodes in the hidden layer of the feed-forward network. The model training and testing were performed with separate data sets of 131 and 100 cases. The concordance correlation coefficient (95% CI) were 0.561 (0.560 to 0.562) in the deep learning model, which was significantly larger than that in the response surface model (0.265 [0.263 to 0.266], P deep learning model-predicted bispectral index during target-controlled infusion of propofol and remifentanil more accurately compared to the traditional model. The deep learning approach in anesthetic pharmacology seems promising because of its excellent performance and extensibility.

  13. Deep learning predictions of survival based on MRI in amyotrophic lateral sclerosis.

    Science.gov (United States)

    van der Burgh, Hannelore K; Schmidt, Ruben; Westeneng, Henk-Jan; de Reus, Marcel A; van den Berg, Leonard H; van den Heuvel, Martijn P

    2017-01-01

    Amyotrophic lateral sclerosis (ALS) is a progressive neuromuscular disease, with large variation in survival between patients. Currently, it remains rather difficult to predict survival based on clinical parameters alone. Here, we set out to use clinical characteristics in combination with MRI data to predict survival of ALS patients using deep learning, a machine learning technique highly effective in a broad range of big-data analyses. A group of 135 ALS patients was included from whom high-resolution diffusion-weighted and T1-weighted images were acquired at the first visit to the outpatient clinic. Next, each of the patients was monitored carefully and survival time to death was recorded. Patients were labeled as short, medium or long survivors, based on their recorded time to death as measured from the time of disease onset. In the deep learning procedure, the total group of 135 patients was split into a training set for deep learning (n = 83 patients), a validation set (n = 20) and an independent evaluation set (n = 32) to evaluate the performance of the obtained deep learning networks. Deep learning based on clinical characteristics predicted survival category correctly in 68.8% of the cases. Deep learning based on MRI predicted 62.5% correctly using structural connectivity and 62.5% using brain morphology data. Notably, when we combined the three sources of information, deep learning prediction accuracy increased to 84.4%. Taken together, our findings show the added value of MRI with respect to predicting survival in ALS, demonstrating the advantage of deep learning in disease prognostication.

  14. Informative sensor selection and learning for prediction of lower limb kinematics using generative stochastic neural networks.

    Science.gov (United States)

    Eunsuk Chong; Taejin Choi; Hyungmin Kim; Seung-Jong Kim; Yoha Hwang; Jong Min Lee

    2017-07-01

    We propose a novel approach of selecting useful input sensors as well as learning a mathematical model for predicting lower limb joint kinematics. We applied a feature selection method based on the mutual information called the variational information maximization, which has been reported as the state-of-the-art work among information based feature selection methods. The main difficulty in applying the method is estimating reliable probability density of input and output data, especially when the data are high dimensional and real-valued. We addressed this problem by applying a generative stochastic neural network called the restricted Boltzmann machine, through which we could perform sampling based probability estimation. The mutual informations between inputs and outputs are evaluated in each backward sensor elimination step, and the least informative sensor is removed with its network connections. The entire network is fine-tuned by maximizing conditional likelihood in each step. Experimental results are shown for 4 healthy subjects walking with various speeds, recording 64 sensor measurements including electromyogram, acceleration, and foot-pressure sensors attached on both lower limbs for predicting hip and knee joint angles. For test set of walking with arbitrary speed, our results show that our suggested method can select informative sensors while maintaining a good prediction accuracy.

  15. Predicting perceptual learning from higher-order cortical processing.

    Science.gov (United States)

    Wang, Fang; Huang, Jing; Lv, Yaping; Ma, Xiaoli; Yang, Bin; Wang, Encong; Du, Boqi; Li, Wu; Song, Yan

    2016-01-01

    Visual perceptual learning has been shown to be highly specific to the retinotopic location and attributes of the trained stimulus. Recent psychophysical studies suggest that these specificities, which have been associated with early retinotopic visual cortex, may in fact not be inherent in perceptual learning and could be related to higher-order brain functions. Here we provide direct electrophysiological evidence in support of this proposition. In a series of event-related potential (ERP) experiments, we recorded high-density electroencephalography (EEG) from human adults over the course of learning in a texture discrimination task (TDT). The results consistently showed that the earliest C1 component (68-84ms), known to reflect V1 activity driven by feedforward inputs, was not modulated by learning regardless of whether the behavioral improvement is location specific or not. In contrast, two later posterior ERP components (posterior P1 and P160-350) over the occipital cortex and one anterior ERP component (anterior P160-350) over the prefrontal cortex were progressively modified day by day. Moreover, the change of the anterior component was closely correlated with improved behavioral performance on a daily basis. Consistent with recent psychophysical and imaging observations, our results indicate that perceptual learning can mainly involve changes in higher-level visual cortex as well as in the neural networks responsible for cognitive functions such as attention and decision making. Copyright © 2015 Elsevier Inc. All rights reserved.

  16. Using Five Machine Learning for Breast Cancer Biopsy Predictions Based on Mammographic Diagnosis

    OpenAIRE

    Oyewola, David; Hakimi, Danladi; Adeboye, Kayode; Shehu, Musa Danjuma

    2017-01-01

    Breast cancer is one of thecauses of female death in the world. Mammography  is commonly  used for  distinguishing  malignant tumors  from benign  ones. In this research,  a mammographic  diagnostic method  is  presented for breast  cancer  biopsy outcome  predictions  using  fivemachine learning which includes: Logistic Regression(LR), Linear DiscriminantAnalysis(LDA), Quadratic Discriminant Analysis(QDA), Random Forest(RF) andSupport  Vector Machine(SVM)  classification.  The testing result...

  17. Spatial extreme learning machines: An application on prediction of disease counts.

    Science.gov (United States)

    Prates, Marcos O

    2018-01-01

    Extreme learning machines have gained a lot of attention by the machine learning community because of its interesting properties and computational advantages. With the increase in collection of information nowadays, many sources of data have missing information making statistical analysis harder or unfeasible. In this paper, we present a new model, coined spatial extreme learning machine, that combine spatial modeling with extreme learning machines keeping the nice properties of both methodologies and making it very flexible and robust. As explained throughout the text, the spatial extreme learning machines have many advantages in comparison with the traditional extreme learning machines. By a simulation study and a real data analysis we present how the spatial extreme learning machine can be used to improve imputation of missing data and uncertainty prediction estimation.

  18. Regulating approaches to learning: Testing learning strategy convergences across a year at university.

    Science.gov (United States)

    Fryer, Luke K; Vermunt, Jan D

    2018-03-01

    Contemporary models of student learning within higher education are often inclusive of processing and regulation strategies. Considerable research has examined their use over time and their (person-centred) convergence. The longitudinal stability/variability of learning strategy use, however, is poorly understood, but essential to supporting student learning across university experiences. Develop and test a person-centred longitudinal model of learning strategies across the first-year university experience. Japanese university students (n = 933) completed surveys (deep and surface approaches to learning; self, external, and lack of regulation) at the beginning and end of their first year. Following invariance and cross-sectional tests, latent profile transition analysis (LPTA) was undertaken. Initial difference testing supported small but significant differences for self-/external regulation. Fit indices supported a four-group model, consistent across both measurement points. These subgroups were labelled Low Quality (low deep approaches and self-regulation), Low Quantity (low strategy use generally), Average (moderate strategy use), and High Quantity (intense use of all strategies) strategies. The stability of these groups ranged from stable to variable: Average (93% stayers), Low Quality (90% stayers), High Quantity (72% stayers), and Low Quantity (40% stayers). The three largest transitions presented joint shifts in processing/regulation strategy preference across the year, from adaptive to maladaptive and vice versa. Person-centred longitudinal findings presented patterns of learning transitions that different students experience during their first year at university. Stability/variability of students' strategy use was linked to the nature of initial subgroup membership. Findings also indicated strong connections between processing and regulation strategy changes across first-year university experiences. Implications for theory and practice are discussed.

  19. A Mobile Health Application to Predict Postpartum Depression Based on Machine Learning.

    Science.gov (United States)

    Jiménez-Serrano, Santiago; Tortajada, Salvador; García-Gómez, Juan Miguel

    2015-07-01

    Postpartum depression (PPD) is a disorder that often goes undiagnosed. The development of a screening program requires considerable and careful effort, where evidence-based decisions have to be taken in order to obtain an effective test with a high level of sensitivity and an acceptable specificity that is quick to perform, easy to interpret, culturally sensitive, and cost-effective. The purpose of this article is twofold: first, to develop classification models for detecting the risk of PPD during the first week after childbirth, thus enabling early intervention; and second, to develop a mobile health (m-health) application (app) for the Android(®) (Google, Mountain View, CA) platform based on the model with best performance for both mothers who have just given birth and clinicians who want to monitor their patient's test. A set of predictive models for estimating the risk of PPD was trained using machine learning techniques and data about postpartum women collected from seven Spanish hospitals. An internal evaluation was carried out using a hold-out strategy. An easy flowchart and architecture for designing the graphical user interface of the m-health app was followed. Naive Bayes showed the best balance between sensitivity and specificity as a predictive model for PPD during the first week after delivery. It was integrated into the clinical decision support system for Android mobile apps. This approach can enable the early prediction and detection of PPD because it fulfills the conditions of an effective screening test with a high level of sensitivity and specificity that is quick to perform, easy to interpret, culturally sensitive, and cost-effective.

  20. Predicting and preventing organizational failure: learning, stability and safety culture

    International Nuclear Information System (INIS)

    Duffey, R.B.

    2009-01-01

    The physical definition of 'safety culture' is the creation of an organizational and operational structure that places unending emphasis on safety at every level. We propose and prefer the use of the term and the objective of sustaining a 'Learning Environment', where mistakes, outcomes and errors are used as learning vehicles to improve, and we can now define why that is true. Therefore we can manage and quantify safety effectively tracking and analyzing outcomes, using the trends to guide our needed organizational behaviors. (author)

  1. Machine learning and predictive data analytics enabling metrology and process control in IC fabrication

    Science.gov (United States)

    Rana, Narender; Zhang, Yunlin; Wall, Donald; Dirahoui, Bachir; Bailey, Todd C.

    2015-03-01

    Integrate circuit (IC) technology is going through multiple changes in terms of patterning techniques (multiple patterning, EUV and DSA), device architectures (FinFET, nanowire, graphene) and patterning scale (few nanometers). These changes require tight controls on processes and measurements to achieve the required device performance, and challenge the metrology and process control in terms of capability and quality. Multivariate data with complex nonlinear trends and correlations generally cannot be described well by mathematical or parametric models but can be relatively easily learned by computing machines and used to predict or extrapolate. This paper introduces the predictive metrology approach which has been applied to three different applications. Machine learning and predictive analytics have been leveraged to accurately predict dimensions of EUV resist patterns down to 18 nm half pitch leveraging resist shrinkage patterns. These patterns could not be directly and accurately measured due to metrology tool limitations. Machine learning has also been applied to predict the electrical performance early in the process pipeline for deep trench capacitance and metal line resistance. As the wafer goes through various processes its associated cost multiplies. It may take days to weeks to get the electrical performance readout. Predicting the electrical performance early on can be very valuable in enabling timely actionable decision such as rework, scrap, feedforward, feedback predicted information or information derived from prediction to improve or monitor processes. This paper provides a general overview of machine learning and advanced analytics application in the advanced semiconductor development and manufacturing.

  2. Automatic stimulation of experiments and learning based on prediction failure recognition

    NARCIS (Netherlands)

    Juarez Cordova, A.G.; Kahl, B.; Henne, T.; Prassler, E.

    2009-01-01

    In this paper we focus on the task of automatically and autonomously initiating experimentation and learning based on the recognition of prediction failure. We present a mechanism that utilizes conceptual knowledge to predict the outcome of robot actions, observes their execution and indicates when

  3. Predicting Academic Success from Academic Motivation and Learning Approaches in Classroom Teaching Students

    Science.gov (United States)

    Çetin, Baris

    2015-01-01

    Our aim was to determine whether learning approaches and academic motivation together predict academic success of classroom teaching students. The sample of the study included 536 students (386 female, 150 male) studying at the Classroom Teaching Division of Canakkale 18 Mart University. Our research was designed as a prediction study. Data was…

  4. Data Exploration and Analysis of Alternative Learning System Accreditation and Equivalency Test Result Using Data Mining

    Science.gov (United States)

    Talingdan, J. A.; Trinidad, J. T., Jr.; Palaoag, T. D.

    2018-03-01

    Alternative Learning System (ALS) is a subsystem of Depatment of Education (DepEd) that serves as an option of learners who cannot afford to go in a formal education. The research focuses on the data exploration and analysis of ALS accreditation and equivalency test result using data mining. The ALS 2014 to 2016 A & E test results in the secondary level were used as data sets in the study. The A & E test results revealed that the passing rate is doubled per year. The results were clustered using k- means clustering algorithm and they were grouped into good, medium, and low standard learners to identify students need exceptional stuff for enhancement. From the clustered data, it was found out that the strand they are weak in is strand 4 which is the Development of Self and a Sense of Community with a general average of 84.23. It also revealed that the essay type of exam got the lowest score with a general average of 2.14 compared to the multiple type of exam that covers the five learning strands. Furthermore, decision tree and naive bayes were also employed in the study to predict the performance of the learners in the A & E test and determine which is better to use for prediction. It was concluded that naive bayes performs better because the accuracy rate is higher than the decision tree algorithm.

  5. Testing the performance of technical trading rules in the Chinese markets based on superior predictive test

    Science.gov (United States)

    Wang, Shan; Jiang, Zhi-Qiang; Li, Sai-Ping; Zhou, Wei-Xing

    2015-12-01

    Technical trading rules have a long history of being used by practitioners in financial markets. The profitable ability and efficiency of technical trading rules are yet controversial. In this paper, we test the performance of more than seven thousand traditional technical trading rules on the Shanghai Securities Composite Index (SSCI) from May 21, 1992 through June 30, 2013 and China Securities Index 300 (CSI 300) from April 8, 2005 through June 30, 2013 to check whether an effective trading strategy could be found by using the performance measurements based on the return and Sharpe ratio. To correct for the influence of the data-snooping effect, we adopt the Superior Predictive Ability test to evaluate if there exists a trading rule that can significantly outperform the benchmark. The result shows that for SSCI, technical trading rules offer significant profitability, while for CSI 300, this ability is lost. We further partition the SSCI into two sub-series and find that the efficiency of technical trading in sub-series, which have exactly the same spanning period as that of CSI 300, is severely weakened. By testing the trading rules on both indexes with a five-year moving window, we find that during the financial bubble from 2005 to 2007, the effectiveness of technical trading rules is greatly improved. This is consistent with the predictive ability of technical trading rules which appears when the market is less efficient.

  6. Traffic Flow Prediction with Rainfall Impact Using a Deep Learning Method

    Directory of Open Access Journals (Sweden)

    Yuhan Jia

    2017-01-01

    Full Text Available Accurate traffic flow prediction is increasingly essential for successful traffic modeling, operation, and management. Traditional data driven traffic flow prediction approaches have largely assumed restrictive (shallow model architectures and do not leverage the large amount of environmental data available. Inspired by deep learning methods with more complex model architectures and effective data mining capabilities, this paper introduces the deep belief network (DBN and long short-term memory (LSTM to predict urban traffic flow considering the impact of rainfall. The rainfall-integrated DBN and LSTM can learn the features of traffic flow under various rainfall scenarios. Experimental results indicate that, with the consideration of additional rainfall factor, the deep learning predictors have better accuracy than existing predictors and also yield improvements over the original deep learning models without rainfall input. Furthermore, the LSTM can outperform the DBN to capture the time series characteristics of traffic flow data.

  7. Enhancing Health Risk Prediction with Deep Learning on Big Data and Revised Fusion Node Paradigm

    Directory of Open Access Journals (Sweden)

    Hongye Zhong

    2017-01-01

    Full Text Available With recent advances in health systems, the amount of health data is expanding rapidly in various formats. This data originates from many new sources including digital records, mobile devices, and wearable health devices. Big health data offers more opportunities for health data analysis and enhancement of health services via innovative approaches. The objective of this research is to develop a framework to enhance health prediction with the revised fusion node and deep learning paradigms. Fusion node is an information fusion model for constructing prediction systems. Deep learning involves the complex application of machine-learning algorithms, such as Bayesian fusions and neural network, for data extraction and logical inference. Deep learning, combined with information fusion paradigms, can be utilized to provide more comprehensive and reliable predictions from big health data. Based on the proposed framework, an experimental system is developed as an illustration for the framework implementation.

  8. Deep architectures and deep learning in chemoinformatics: the prediction of aqueous solubility for drug-like molecules.

    Science.gov (United States)

    Lusci, Alessandro; Pollastri, Gianluca; Baldi, Pierre

    2013-07-22

    Shallow machine learning methods have been applied to chemoinformatics problems with some success. As more data becomes available and more complex problems are tackled, deep machine learning methods may also become useful. Here, we present a brief overview of deep learning methods and show in particular how recursive neural network approaches can be applied to the problem of predicting molecular properties. However, molecules are typically described by undirected cyclic graphs, while recursive approaches typically use directed acyclic graphs. Thus, we develop methods to address this discrepancy, essentially by considering an ensemble of recursive neural networks associated with all possible vertex-centered acyclic orientations of the molecular graph. One advantage of this approach is that it relies only minimally on the identification of suitable molecular descriptors because suitable representations are learned automatically from the data. Several variants of this approach are applied to the problem of predicting aqueous solubility and tested on four benchmark data sets. Experimental results show that the performance of the deep learning methods matches or exceeds the performance of other state-of-the-art methods according to several evaluation metrics and expose the fundamental limitations arising from training sets that are too small or too noisy. A Web-based predictor, AquaSol, is available online through the ChemDB portal ( cdb.ics.uci.edu ) together with additional material.

  9. Sensor Data Air Pollution Prediction by Machine Learning Methods

    Czech Academy of Sciences Publication Activity Database

    Vidnerová, Petra; Neruda, Roman

    submitted 25. 1. (2018) ISSN 1530-437X R&D Projects: GA ČR GA15-18108S Grant - others:GA MŠk(CZ) LM2015042 Institutional support: RVO:67985807 Keywords : machine learning * sensors * air pollution * deep neural networks * regularization networks Subject RIV: IN - Informatics, Computer Science Impact factor: 2.512, year: 2016

  10. Deep learning for predicting the monsoon over the homogeneous ...

    Indian Academy of Sciences (India)

    Moumita Saha

    2017-06-12

    Jun 12, 2017 ... governs the pulse of life for its mankind along with the flora-fauna ..... work. The architecture has the ability to learn the properties of input data space, and it is utilized for feature .... motivation of minimizing the reconstruction error of the input (the .... balance between the accuracy and speed of the method.

  11. Learning in manufacturing organizations : what factors predict effectiveness?

    NARCIS (Netherlands)

    Shipton, H.; Dawson, J.; West, M.A.; Patterson, M.G.

    2002-01-01

    This paper argues that it is possible to identify factors which pre-dispose organizations to adopt effective learning strategies and processes. It is hypothesized that effective OL is associated with: profitability, environmental uncertainty, structure, approach to HRM and quality orientation. The

  12. Competitive Learning Neural Network Ensemble Weighted by Predicted Performance

    Science.gov (United States)

    Ye, Qiang

    2010-01-01

    Ensemble approaches have been shown to enhance classification by combining the outputs from a set of voting classifiers. Diversity in error patterns among base classifiers promotes ensemble performance. Multi-task learning is an important characteristic for Neural Network classifiers. Introducing a secondary output unit that receives different…

  13. A Long Short-Term Memory deep learning network for the prediction of epileptic seizures using EEG signals.

    Science.gov (United States)

    Tsiouris, Κostas Μ; Pezoulas, Vasileios C; Zervakis, Michalis; Konitsiotis, Spiros; Koutsouris, Dimitrios D; Fotiadis, Dimitrios I

    2018-05-17

    The electroencephalogram (EEG) is the most prominent means to study epilepsy and capture changes in electrical brain activity that could declare an imminent seizure. In this work, Long Short-Term Memory (LSTM) networks are introduced in epileptic seizure prediction using EEG signals, expanding the use of deep learning algorithms with convolutional neural networks (CNN). A pre-analysis is initially performed to find the optimal architecture of the LSTM network by testing several modules and layers of memory units. Based on these results, a two-layer LSTM network is selected to evaluate seizure prediction performance using four different lengths of preictal windows, ranging from 15 min to 2 h. The LSTM model exploits a wide range of features extracted prior to classification, including time and frequency domain features, between EEG channels cross-correlation and graph theoretic features. The evaluation is performed using long-term EEG recordings from the open CHB-MIT Scalp EEG database, suggest that the proposed methodology is able to predict all 185 seizures, providing high rates of seizure prediction sensitivity and low false prediction rates (FPR) of 0.11-0.02 false alarms per hour, depending on the duration of the preictal window. The proposed LSTM-based methodology delivers a significant increase in seizure prediction performance compared to both traditional machine learning techniques and convolutional neural networks that have been previously evaluated in the literature. Copyright © 2018 Elsevier Ltd. All rights reserved.

  14. Understanding and Predicting Student Self-Regulated Learning Strategies in Game-Based Learning Environments

    Science.gov (United States)

    Sabourin, Jennifer L.; Shores, Lucy R.; Mott, Bradford W.; Lester, James C.

    2013-01-01

    Self-regulated learning behaviors such as goal setting and monitoring have been found to be crucial to students' success in computer-based learning environments. Consequently, understanding students' self-regulated learning behavior has been the subject of increasing attention. Unfortunately, monitoring these behaviors in real-time has…

  15. Interaction of Instrumental and Goal-Directed Learning Modulates Prediction Error Representations in the Ventral Striatum.

    Science.gov (United States)

    Guo, Rong; Böhmer, Wendelin; Hebart, Martin; Chien, Samson; Sommer, Tobias; Obermayer, Klaus; Gläscher, Jan

    2016-12-14

    Goal-directed and instrumental learning are both important controllers of human behavior. Learning about which stimulus event occurs in the environment and the reward associated with them allows humans to seek out the most valuable stimulus and move through the environment in a goal-directed manner. Stimulus-response associations are characteristic of instrumental learning, whereas response-outcome associations are the hallmark of goal-directed learning. Here we provide behavioral, computational, and neuroimaging results from a novel task in which stimulus-response and response-outcome associations are learned simultaneously but dominate behavior at different stages of the experiment. We found that prediction error representations in the ventral striatum depend on which type of learning dominates. Furthermore, the amygdala tracks the time-dependent weighting of stimulus-response versus response-outcome learning. Our findings suggest that the goal-directed and instrumental controllers dynamically engage the ventral striatum in representing prediction errors whenever one of them is dominating choice behavior. Converging evidence in human neuroimaging studies has shown that the reward prediction errors are correlated with activity in the ventral striatum. Our results demonstrate that this region is simultaneously correlated with a stimulus