{"id":"W4297312826","doi":"10.3389/fdata.2022.979465","title":"The ABC recommendations for validation of supervised machine learning results in biomedical sciences","year":2022,"lang":"en","type":"article","venue":"Frontiers in Big Data","topic":"Explainable Artificial Intelligence (XAI)","field":"Computer Science","cited_by":14,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"","keywords":"Big data; Front (military); Public opinion; Volume (thermodynamics); Data science; Public health; Artificial intelligence; Computer science; Medicine; Political science; Engineering; Data mining; Pathology; Mechanical engineering","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":[0.004681274,0.00007058057,0.0001203565,0.0003242452,0.0006289316,0.00009193583,0.002979943,0.00002429127,0.000005018198],"category_scores_gemma":[0.00125265,0.00006078115,0.00001942433,0.001537837,0.0001820461,0.0004963463,0.001233002,0.0002083692,0.000001253958],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00008585957,"about_ca_system_score_gemma":0.0001590673,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0004373878,"about_ca_topic_score_gemma":0.0001864678,"domain_scores_codex":[0.9981838,0.0003202444,0.0004614247,0.0004287552,0.0003366263,0.0002691707],"domain_scores_gemma":[0.9986029,0.0004786617,0.0001387611,0.0007140477,0.00003415317,0.00003152722],"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":[0.0001839051,0.0003730589,0.01255088,0.00002421692,0.00002021294,0.000007653136,0.005367198,0.008126047,0.0005325868,0.01165786,0.1158036,0.8453527],"study_design_scores_gemma":[0.0003028631,0.0001562054,0.0001402969,0.00001445808,0.000002029361,0.000001282174,0.002276493,0.7900706,0.001098943,0.006944322,0.1988988,0.00009365407],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.009705001,0.0005716364,0.954067,0.03009912,0.00402509,0.0006377345,0.000529512,0.00005318449,0.0003117516],"genre_scores_gemma":[0.7276583,0.0003826793,0.2693042,0.0002476783,0.0001214922,0.0001885454,0.001777924,0.00001608515,0.0003031381],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.8452591,"threshold_uncertainty_score":0.5537524,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1559307450808685,"score_gpt":0.3363171335948982,"score_spread":0.1803863885140297,"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."}}