{"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":"metacan-v3-hybrid-931329e0061c","candidate_categories":["metaresearch"],"consensus_categories":["metaresearch"],"category_scores_codex":[0.7485245,0.003738106,0.008288046,0.01497985,0.004720542,0.02063948,0.0125633,0.01169211,0.001935266],"category_scores_gemma":[0.8559234,0.004589535,0.009443387,0.01076754,0.03284246,0.01567928,0.01784046,0.01354838,0.001258952],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.006784667,"about_ca_system_score_gemma":0.01893363,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00375261,"about_ca_topic_score_gemma":0.001941402,"domain_scores_codex":[0.2196726,0.6555229,0.059662,0.0193329,0.04429592,0.001513756],"domain_scores_gemma":[0.07987326,0.7989997,0.02859594,0.06280734,0.02880435,0.0009194126],"domain_codex":"methods","domain_gemma":"methods","domain_candidate":"methods","domain_consensus":"methods","study_design_codex":"theoretical_or_conceptual","study_design_gemma":"theoretical_or_conceptual","study_design_scores_codex":[0.001693241,0.0005240602,0.04750939,0.0126821,0.007404794,0.0005742051,0.008844757,0.04735822,0.001514254,0.6042644,0.008504163,0.2591265],"study_design_scores_gemma":[0.001429799,0.001085871,0.009263342,0.01359835,0.001406996,0.000830101,0.0008367818,0.08318125,0.004450747,0.856177,0.02735681,0.0003829766],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":"theoretical_or_conceptual","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.003465902,0.003566727,0.9822411,0.003553408,0.0003856845,0.004197042,0.0002834764,0.0002587988,0.002047864],"genre_scores_gemma":[0.04204156,0.001196253,0.9451894,0.001613473,0.0004618821,0.008815587,0.0003046267,0.00014797,0.0002290843],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.2514755,"threshold_uncertainty_score":0.3101141,"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."}}