{"id":"W4381487215","doi":"10.1016/j.jbi.2023.104426","title":"A statistical comparison between Matthews correlation coefficient (MCC), prevalence threshold, and Fowlkes–Mallows index","year":2023,"lang":"en","type":"article","venue":"Journal of Biomedical Informatics","topic":"Imbalanced Data Classification Techniques","field":"Computer Science","cited_by":48,"is_retracted":false,"has_abstract":false,"ca_institutions":"University of Toronto","funders":"","keywords":"Statistics; Correlation coefficient; Cohen's kappa; Correlation; Statistic; Brier score; Kappa; Mathematics; Computer science; Data mining; Econometrics","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":[],"consensus_categories":[],"category_scores_codex":[0.001571448,0.0001369034,0.0003374298,0.0004427839,0.0001010342,0.0001932495,0.0008073613,0.0001413437,0.00001451062],"category_scores_gemma":[0.0003556814,0.0001069242,0.00004547893,0.0007334898,0.0002533311,0.0008106807,0.0003145054,0.0003050666,0.00005096027],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00007200355,"about_ca_system_score_gemma":0.000136705,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000001021608,"about_ca_topic_score_gemma":2.537428e-7,"domain_scores_codex":[0.9971308,0.00005626648,0.001406649,0.0001040861,0.001026366,0.0002758462],"domain_scores_gemma":[0.9979228,0.0004500952,0.0008274393,0.0003263141,0.0001913468,0.0002820511],"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":[0.00008692213,0.0006225736,0.2085175,0.001500644,0.0002214556,0.00007992755,0.01363252,0.0007693158,0.0005271047,0.05092364,0.3054557,0.4176627],"study_design_scores_gemma":[0.0005958104,0.0004344309,0.1328337,0.0002063298,0.00002850409,0.00008576199,0.0002844984,0.845311,0.0001366304,0.002725428,0.01716202,0.0001958705],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.03546181,0.00003528599,0.9629496,0.0008410544,0.0002610543,0.0001556444,0.00004121286,0.0001296571,0.000124657],"genre_scores_gemma":[0.8800969,0.0001408879,0.1193508,0.0002223526,0.00009408479,0.000005177051,0.00004394972,0.000008541653,0.0000373145],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.8446351,"threshold_uncertainty_score":0.4360242,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03000122756346132,"score_gpt":0.3158772012061198,"score_spread":0.2858759736426585,"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."}}