640 Hits in 2.5 sec

Fair Performance Metric Elicitation [article]

Gaurush Hiranandani, Harikrishna Narasimhan, Oluwasanmi Koyejo
2020 arXiv   pre-print
Specifically, we propose a novel strategy to elicit group-fair performance metrics for multiclass classification problems with multiple sensitive groups that also includes selecting the trade-off between  ...  The use of metric elicitation enables a practitioner to tune the performance and fairness metrics to the task, context, and population at hand.  ...  To this end, we extend the ME framework from eliciting multiclass classification metrics [21] to the task of eliciting fair performance metrics from pairwise preference feedback in the presence of multiple  ... 
arXiv:2006.12732v3 fatcat:74dfxqqdn5butnw3nbxx3iyzca

Quadratic Metric Elicitation for Fairness and Beyond [article]

Gaurush Hiranandani, Jatin Mathur, Harikrishna Narasimhan, Oluwasanmi Koyejo
2021 arXiv   pre-print
Metric elicitation is a recent framework for eliciting performance metrics that best reflect implicit user preferences based on the application and context.  ...  This paper develops a strategy for eliciting more flexible multiclass metrics defined by quadratic functions of rates, designed to reflect human preferences better.  ...  Appendix for "Quadratic Metric Elicitation for Fairness and Beyond" A Linear Performance Metric Elicitation (LPME) In this section, we shed more light on the procedure from [19] that elicits a multiclass  ... 
arXiv:2011.01516v2 fatcat:ew2rk6u4jzbhzcwxr6sj2677ma

Estimating smile intensity: A better way

Jeffrey M. Girard, Jeffrey F. Cohn, Fernando De la Torre
2015 Pattern Recognition Letters  
Multiclass models even outperformed binarytrained classifiers on smile occurrence detection.  ...  The most straight-forward approach is to train multiclass or regression models using intensity ground truth.  ...  This property is desirable because many performance metrics are highly sensitive to class skew (Jeni et al., 2013a) .  ... 
doi:10.1016/j.patrec.2014.10.004 pmid:26461205 pmcid:PMC4598946 fatcat:qmdajjlj3fbfnamyor2ywa7cwu

Bayesian Variable Selection for Multiclass Classification using Bootstrap Prior Technique

Oyebayo Ridwan Olaniran, Mohd Asrul Affendi Abdullah
2019 Austrian Journal of Statistics  
The test function developed was later used for variable screening in multiclass classification scenario.  ...  Performance comparison between the proposed method and existing classical ANOVA method was achieved using simulated and real life gene expression datasets.  ...  Variable selection in multiclass classification is the process of identifying relevant subset of input variables that can positively improve the performance of a multiclass classifier.  ... 
doi:10.17713/ajs.v48i2.806 fatcat:v6rkil4oevhmhfqy3h56swfbfu

Prediction of Human Empathy based on EEG Cortical Asymmetry [article]

Andrea Kuijt, Maryam Alimardani
2020 arXiv   pre-print
Different types of predictive models i.e. multiple linear regression analyses as well as binary and multiclass classifications were evaluated.  ...  Additionally, prominent classification performance was found during resting-state which suggests that emotional stimulation is not required for accurate prediction of empathy -- as a personality trait  ...  When comparing the performance of the binary classification models with the performance of the multiclass classification models, better results were found for the binary classification than for the multiclass  ... 
arXiv:2005.02824v1 fatcat:oeqq5se4k5ccbpwrlxpo6rmp2m

The binomial-neighbour instance-based learner on a multiclass performance measure scheme

Theodoros Theodoridis, Huosheng Hu
2014 Soft Computing - A Fusion of Foundations, Methodologies and Applications  
The paper introduces as well a performance measure scheme for multiclass problems using type error statistics.  ...  Classification results are being compared with the standard IBk and IB1 models achieving significantly exceptional recognition performance.  ...  The original statistical functions derived from performance metrics, which have been used for the multiclass problem, are included in the work of Sokolova et al. (2006) .  ... 
doi:10.1007/s00500-014-1461-z fatcat:nfl5h5zr3vgnxbqfaece6deaai

Machine Learning of Human Pluripotent Stem Cell-Derived Engineered Cardiac Tissue Contractility for Automated Drug Classification

Eugene K. Lee, David D. Tran, Wendy Keung, Patrick Chan, Gabriel Wong, Camie W. Chan, Kevin D. Costa, Ronald A. Li, Michelle Khine
2017 Stem Cell Reports  
A generated metric is effective for then determining the cardioactivity of a given drug.  ...  In all three conditions, the multiclass models behaved similarly in that both the nifedipine and isoproterenol classifiers performed the best by always achieving the highest F 1 score values, a metric  ...  (Right) Metrics indicate a similar performance between conditions 1 and 2. (C) Condition 3: flecainide and E-4031.  ... 
doi:10.1016/j.stemcr.2017.09.008 pmid:29033305 pmcid:PMC5829317 fatcat:y7jnm2kmkjcdfjshnjfiitr2oy

Supporting End-User Understanding of Classification Errors: Visualization and Usability Issues

Emma M.A.L. Beauxis-Aussalet, Joost Van Doorn, Lynda Hardman
2019 Journal of Interaction Science  
The end-user-oriented visualization reduced the difficulties by using several visual features to clarify the actual and predicted classes, and more tangible metrics and representation.  ...  Fig. 11 . 11 ROC curves used for binary and multiclass data. Fig. 12 . 12 Confusion matrices used for tasks T2-7 to T2-9. Fig. 14 . 14 Task performance per visualization.  ...  More detailed and generalizable insights on the usability are elicited from our qualitative analysis of user interviews.  ... 
doi:10.24982/jois.1814019.003 fatcat:ticsxzvxfnfgliafzsuiabpfxu

A predictive model for network intrusion detection using stacking approach

Smitha Rajagopal, Poornima Panduranga Kundapur, Hareesh Katiganere Siddaramappa
2020 International Journal of Electrical and Computer Engineering (IJECE)  
Empirical investigation has illustrated that the performance of the proposed approach has been reasonably good.  ...  Therefore, the proposed work on network intrusion detection emphasizes upon a combinative approach to improve performance.  ...  It is a common practice in machine learning research to illustrate the performance of a classifier by considering various metrics since each metric holds its own relevance.  ... 
doi:10.11591/ijece.v10i3.pp2734-2741 fatcat:42kjlmklsbfonpgthvlfji5qaq

Chronic Kidney Disease Prediction using Machine Learning Models

2019 International Journal of Engineering and Advanced Technology  
, Multiclass Decision Forest, Multiclass Neural Network and Multiclass Logistic Regression.  ...  Datamining methods are used to generate decisions by elicitating hidden information from chronic disease datasets.  ... 
doi:10.35940/ijeat.a2213.109119 fatcat:62ofj5lkazds5okz5i4634ucjq

Predicting antimicrobial mechanism-of-action from transcriptomes: A generalizable explainable artificial intelligence approach

Josh L. Espinoza, Chris L. Dupont, Aubrie O'Rourke, Sinem Beyhan, Pavel Morales, Amy Spoering, Kirsten J. Meyer, Agnes P. Chan, Yongwook Choi, William C. Nierman, Kim Lewis, Karen E. Nelson (+1 others)
2021 PLoS Computational Biology  
; improving upon the performance metrics of the original publications.  ...  Here we describe an explainable artificial intelligence classification methodology that emphasizes prediction performance and human interpretability by using a Hierarchical Ensemble of Classifiers model  ...  Performance metrics for each LCOCV set include accuracy, precision, recall, and F1 score.  ... 
doi:10.1371/journal.pcbi.1008857 pmid:33780444 fatcat:747cakuwj5hbtbyyuasecm6vnq

Sentiment Analysis by Fusing Text and Location Features of Geo-tagged Tweets

Wei Lun Lim, Chiung Ching Ho, Choo-Yee Ting
2020 IEEE Access  
At the same time, well-kept natural surroundings elicited strong positive sentiments [10] . There is a correlation between urban metrics and sentiment analysis.  ...  The experiments were performed on four categories with two on binary classification and two on multiclass classification.  ...  embedding approaches like node2vec and random walk to generate the embedding. 5) Include more properties to location data such as distance between a tweet to a location. 6) Use graph neural network to perform  ... 
doi:10.1109/access.2020.3027845 fatcat:wdd2mzup2zbevjxuo4o6ckq3kq

Brain-Machine Interface for Mechanical Ventilation Using Respiratory-Related Evoked Potential [chapter]

Sylvain Chevallier, Guillaume Bao, Mayssa Hammami, Fabienne Marlats, Louis Mayaud, Djillali Annane, Frédéric Lofaso, Eric Azabou
2018 Lecture Notes in Computer Science  
a multiclass detection of the respiratory load, -the obtained results outperform previously reported results, in a more challenging setup (multiclass instead of two classes).  ...  It is possible to choose a metric such that the inner product on the tangent space T Σ M of each point Σ varies smoothly from one point to another.  ... 
doi:10.1007/978-3-030-01424-7_65 fatcat:4epbhex4pvccxjix6cg6sngvgm

Optimizing Black-box Metrics with Iterative Example Weighting [article]

Gaurush Hiranandani, Jatin Mathur, Harikrishna Narasimhan, Mahdi Milani Fard, Oluwasanmi Koyejo
2021 arXiv   pre-print
evaluate the metric via performance evaluation using a small validation sample.  ...  We consider learning to optimize a classification metric defined by a black-box function of the confusion matrix.  ...  Multiclass performance metric elicitation. In Advances in Neural Information Processing Systems, pages 9351-9360, 2019b.  ... 
arXiv:2102.09492v2 fatcat:pji2kmya5fejflcaaaayeeladm

Automatic Detection of Compensation During Robotic Stroke Rehabilitation Therapy

Ying Xuan Zhi, Michelle Lukasik, Michael H. Li, Elham Dolatabadi, Rosalie H. Wang, Babak Taati
2018 IEEE Journal of Translational Engineering in Health and Medicine  
The 3-D trajectories of upper body joint positions tracked over time were used for multiclass classification of postures.  ...  The Receiver Operating Characteristic (ROC) and F1-score for each class were used as performance metrics.  ...  MULTICLASS CLASSIFICATION A multiclass linear Support Vector Machine (SVM) classifier and a Recurrent Neural Network (RNN) classifier were trained to classify the posture of the participants.  ... 
doi:10.1109/jtehm.2017.2780836 pmid:29404226 pmcid:PMC5788403 fatcat:vkrfd36y7jdmvlvgieqrsfvus4
« Previous Showing results 1 — 15 out of 640 results