{"id":"W4397022047","doi":"10.1093/aje/kwae071","title":"Validation of algorithms in studies based on routinely collected health data: general principles","year":2024,"lang":"en","type":"article","venue":"American Journal of Epidemiology","topic":"Advanced Causal Inference Techniques","field":"Mathematics","cited_by":39,"is_retracted":false,"has_abstract":true,"ca_institutions":"Institute for Clinical Evaluative Sciences; SickKids Foundation; University of Toronto","funders":"Galderma; National Institutes of Health; Aarhus Universitet; Region Midtjylland; Sanofi","keywords":"Computer science; Operationalization; External validity; Data mining; Terminology; Algorithm; Context (archaeology); Covariate; Data science; Management science; Machine learning; Statistics; Mathematics","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":["metaresearch"],"consensus_categories":[],"category_scores_codex":[0.006730373,0.0001549016,0.001265655,0.0004121824,0.00002524091,0.000003749471,0.0003240582,0.00004427025,0.0000107786],"category_scores_gemma":[0.01664904,0.000114323,0.00008140888,0.0005504315,0.0003376793,0.0001309915,0.00007907944,0.0003778333,8.633476e-7],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002621132,"about_ca_system_score_gemma":0.0003685968,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001064843,"about_ca_topic_score_gemma":0.00003014978,"domain_scores_codex":[0.9961113,0.001509839,0.001709831,0.0002399747,0.0001479574,0.000281075],"domain_scores_gemma":[0.9879976,0.009818053,0.001537577,0.0003792451,0.0001942164,0.00007331472],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"theoretical_or_conceptual","study_design_scores_codex":[0.001573775,0.001674736,0.07952616,0.001836025,0.001904058,0.0004209785,0.005583978,0.05276581,0.003801354,0.454784,0.0680524,0.3280768],"study_design_scores_gemma":[0.001732581,0.01772599,0.01492341,0.005051505,0.0002321264,0.0004044255,0.002672311,0.235667,0.004791612,0.7082838,0.007587171,0.0009280771],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":"theoretical_or_conceptual","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.3340271,0.002499331,0.6518561,0.01039482,0.0004393568,0.0003922842,0.00008640785,0.0001223768,0.000182299],"genre_scores_gemma":[0.4195755,0.0009277157,0.578496,0.0007737041,0.0001440011,0.000008147227,0.00001770929,0.00002438761,0.00003276294],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.3271486,"threshold_uncertainty_score":0.9916341,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.48073120424196,"score_gpt":0.5450480657461021,"score_spread":0.06431686150414212,"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."}}