{"id":"W3172444956","doi":"10.1109/access.2021.3084050","title":"The Matthews Correlation Coefficient (MCC) is More Informative Than Cohen’s Kappa and Brier Score in Binary Classification Assessment","year":2021,"lang":"en","type":"article","venue":"IEEE Access","topic":"Reliability and Agreement in Measurement","field":"Decision Sciences","cited_by":429,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"","keywords":"Brier score; Kappa; Cohen's kappa; Correlation; Computer science; Statistics; Artificial intelligence; Binary classification; Metric (unit); Binary number; Odds; Machine learning; Mathematics; Natural language processing; Logistic regression; Support vector machine; Arithmetic","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["scholarly_communication"],"consensus_categories":[],"category_scores_codex":[0.003729351,0.0001569236,0.0002169389,0.0001314416,0.0004161778,0.001092042,0.000685083,0.0000848515,0.0001510596],"category_scores_gemma":[0.0004900451,0.00009778181,0.00006522963,0.0008687254,0.0001973012,0.0008947063,0.0002116759,0.0002486232,0.0000557343],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000169706,"about_ca_system_score_gemma":0.0001908761,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00006648133,"about_ca_topic_score_gemma":0.0002266232,"domain_scores_codex":[0.9964032,0.0003539039,0.0008376134,0.0004393026,0.001714478,0.000251458],"domain_scores_gemma":[0.9974095,0.0009074268,0.0003736406,0.0006763844,0.0005540413,0.00007899453],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.00007983487,0.0003187938,0.906014,0.0000313883,0.00003205771,0.000009942843,0.006527978,0.008035145,0.0009332365,0.001957368,0.02247477,0.05358554],"study_design_scores_gemma":[0.0004122015,0.00003853438,0.9109122,0.00006662534,0.0000103665,0.000003421452,0.004770107,0.06985734,0.001062906,0.004363029,0.008355946,0.0001472793],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9791437,0.0001607185,0.008619796,0.007217608,0.0008075006,0.0005620633,0.00001814902,0.00001726408,0.00345317],"genre_scores_gemma":[0.9980797,0.0001387113,0.0001284418,0.0008357337,0.00003697666,0.00007367897,0.00001260196,0.000006193337,0.0006879289],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.0618222,"threshold_uncertainty_score":0.9999449,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1877563290803927,"score_gpt":0.4289589341158113,"score_spread":0.2412026050354187,"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."}}