{"id":"W4385144251","doi":"10.2196/43638","title":"Large Language Models for Epidemiological Research via Automated Machine Learning: Case Study Using Data From the British National Child Development Study","year":2023,"lang":"en","type":"article","venue":"JMIR Medical Informatics","topic":"Mental Health via Writing","field":"Psychology","cited_by":6,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Steno Diabetes Center Aarhus; Novo Nordisk; Novo Nordisk Fonden; Steno Diabetes Center Copenhagen","keywords":"Computer science; Data science; Artificial intelligence; Natural language processing; Machine learning","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":true,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.02104557,0.0007793639,0.0007728592,0.001904743,0.001145296,0.001789339,0.001581075,0.00137903,0.001790549],"category_scores_gemma":[0.07537356,0.000547491,0.00104074,0.002872744,0.001013417,0.001288397,0.00154996,0.002524981,0.0006838372],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00261207,"about_ca_system_score_gemma":0.002122866,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.09057856,"about_ca_topic_score_gemma":0.1083998,"domain_scores_codex":[0.9824461,0.01528615,0.0005628221,0.0007794939,0.0006736452,0.0002517924],"domain_scores_gemma":[0.8136315,0.1742783,0.002694164,0.004814262,0.00373921,0.0008424976],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0009994763,0.001971331,0.746339,0.0007612833,0.0007254417,0.004196799,0.007657181,0.08969456,0.0007546025,0.003492103,0.01493022,0.1284781],"study_design_scores_gemma":[0.0003478882,0.0005895836,0.2871768,0.0005925788,0.0003715851,0.001818541,0.006996156,0.6797554,0.001258257,0.00801615,0.01292062,0.0001563954],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9750875,0.001098443,0.01682757,0.002307261,0.00003628741,0.0002913995,0.002649307,0.0002274262,0.001474863],"genre_scores_gemma":[0.9710686,0.0003898078,0.02300967,0.000192137,0.00004965733,0.0003159732,0.00437772,0.00006446256,0.0005319451],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.09057856,"threshold_uncertainty_score":0.1801026,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.3415056858925381,"score_gpt":0.5482106378465419,"score_spread":0.2067049519540038,"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."}}