{"id":"W2999309192","doi":"10.1186/s12864-019-6413-7","title":"The advantages of the Matthews correlation coefficient (MCC) over F1 score and accuracy in binary classification evaluation","year":2020,"lang":"en","type":"article","venue":"BMC Genomics","topic":"Imbalanced Data Classification Techniques","field":"Computer Science","cited_by":5819,"is_retracted":false,"has_abstract":true,"ca_institutions":"Ontario Tobacco Research Unit; Krembil Foundation","funders":"University of Toronto","keywords":"Binary classification; False positive paradox; Binary number; False positives and false negatives; Correlation; Confusion matrix; Artificial intelligence; Statistics; Pearson product-moment correlation coefficient; False positive rate; Confusion; Correlation coefficient; Matthews correlation coefficient; Computer science; Machine learning; Pattern recognition (psychology); Data mining; Mathematics; Support vector machine; Psychology; Arithmetic","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":[0.02849796,0.001570788,0.001730638,0.0132768,0.001649264,0.003726947,0.001233387,0.002016367,0.001411519],"category_scores_gemma":[0.1234165,0.0002962744,0.0009700214,0.008086227,0.002699645,0.002936627,0.001575349,0.001475063,0.0007575343],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002163354,"about_ca_system_score_gemma":0.001871885,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.005362286,"about_ca_topic_score_gemma":0.005485408,"domain_scores_codex":[0.9655034,0.0136621,0.003428197,0.003981269,0.01238449,0.001040683],"domain_scores_gemma":[0.8101692,0.1292816,0.01820082,0.009798056,0.02989656,0.002653767],"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":[0.002108999,0.0003611165,0.2894453,0.001849624,0.001746795,0.0007493255,0.001593378,0.06926706,0.01610241,0.03856149,0.04058396,0.5376304],"study_design_scores_gemma":[0.0001214639,0.002223496,0.2467224,0.0008546394,0.0007090099,0.00225738,0.001086595,0.613002,0.04941323,0.06159964,0.02128135,0.0007288244],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"methods","genre_scores_codex":[0.5168672,0.01470315,0.4227583,0.00447984,0.001322152,0.0006038364,0.004069156,0.005361022,0.02983544],"genre_scores_gemma":[0.9229698,0.0006912611,0.07248671,0.0003321089,0.0004510386,0.0002446868,0.00111687,0.0003675778,0.001339898],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.02849796,"threshold_uncertainty_score":0.1507134,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.06413411072449375,"score_gpt":0.3027767276479773,"score_spread":0.2386426169234835,"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."}}