{"id":"W4286009680","doi":"10.1101/2022.07.18.500262","title":"Class imbalance should not throw you off balance: Choosing the right classifiers and performance metrics for brain decoding with imbalanced data","year":2022,"lang":"en","type":"preprint","venue":"bioRxiv (Cold Spring Harbor Laboratory)","topic":"Imbalanced Data Classification Techniques","field":"Computer Science","cited_by":13,"is_retracted":false,"has_abstract":true,"ca_institutions":"Mila - Quebec Artificial Intelligence Institute; McGill University; Concordia University; University of Alberta; Université de Montréal","funders":"Fonds de recherche du Québec – Nature et technologies; Courtois Foundation; Natural Sciences and Engineering Research Council of Canada; Institut de Valorisation des Données; Canada Research Chairs; Canada First Research Excellence Fund; Canadian Institutes of Health Research; Mitacs; McGill University","keywords":"Computer science; Magnetoencephalography; Metric (unit); Artificial intelligence; Robustness (evolution); Binary classification; Machine learning; Decoding methods; Random forest; Class (philosophy); Receiver operating characteristic; Performance metric; Electroencephalography; Binary number; Sensitivity (control systems); Pattern recognition (psychology); Support vector machine; Mathematics; Psychology; Algorithm","routes":{"ca_aff":true,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.01564852,0.001291504,0.001105412,0.002539915,0.0006596957,0.002621198,0.0009527698,0.001632734,0.0008631381],"category_scores_gemma":[0.06516194,0.0002393585,0.0004674733,0.001580672,0.001372134,0.002912848,0.001655207,0.002146736,0.0005793593],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00113794,"about_ca_system_score_gemma":0.001160512,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002003905,"about_ca_topic_score_gemma":0.001474109,"domain_scores_codex":[0.9940875,0.003354349,0.0003912619,0.0008936774,0.0009402373,0.0003329482],"domain_scores_gemma":[0.9728505,0.01911916,0.002418067,0.002006574,0.002874241,0.0007314045],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.003165498,0.000688856,0.1070872,0.0006744788,0.0006486675,0.0005864804,0.0009518514,0.3975581,0.02473118,0.01769295,0.01869605,0.4275186],"study_design_scores_gemma":[0.00005211084,0.0003816737,0.01153816,0.0001681815,0.00006526578,0.000205294,0.0003224819,0.9331462,0.01432041,0.03794438,0.001779835,0.00007611309],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.4749351,0.003625431,0.5116073,0.003283435,0.0004531674,0.0002757971,0.0009674498,0.001521999,0.003330343],"genre_scores_gemma":[0.9267815,0.0003188922,0.07117651,0.0002889839,0.0001263747,0.0001213266,0.000694536,0.0001773009,0.0003145574],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.01564852,"threshold_uncertainty_score":0.08275825,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04476784390055522,"score_gpt":0.2714744256389494,"score_spread":0.2267065817383942,"validation_status":"score_only:v0-immature-baseline","note":"Baseline scores from an immature model (maturity gate not passed). Scores rank; they never assert a category."}}