{"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":"codex-gemma-dda1882f352a","candidate_categories":["sts"],"consensus_categories":[],"category_scores_codex":[0.01802718,0.0001879716,0.0003915702,0.0001203854,0.001990015,0.0001464312,0.0009876514,0.0002216211,0.0005259993],"category_scores_gemma":[0.002261294,0.0001458592,0.00003720227,0.0006327286,0.000111646,0.0002424582,0.001441616,0.001377644,0.0002407466],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000167287,"about_ca_system_score_gemma":0.0002397163,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003184924,"about_ca_topic_score_gemma":0.003010916,"domain_scores_codex":[0.9942785,0.001320269,0.00143353,0.0003490851,0.001745547,0.000873096],"domain_scores_gemma":[0.9949982,0.003718996,0.0002376167,0.0005554615,0.0001718108,0.0003179048],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"qualitative","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00037405,0.01155516,0.1125248,0.0005960825,0.001219374,0.01027153,0.7061487,0.0007774637,0.000001779984,0.0004007019,0.1070615,0.0490689],"study_design_scores_gemma":[0.002124656,0.0002101369,0.006903018,0.00009968224,0.00001206043,0.0005740435,0.2324955,0.7561048,1.623282e-7,0.00006419793,0.001276101,0.0001356182],"study_design_candidate":"qualitative","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9922085,0.00009542834,0.002277583,0.0002803339,0.000225486,0.003405437,0.0003284889,0.0005247546,0.0006539946],"genre_scores_gemma":[0.9947226,0.000003839933,0.001633301,0.000962295,0.000279899,0.0005854129,0.001669616,0.00003066477,0.0001123579],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.7553273,"threshold_uncertainty_score":0.9993092,"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."}}