{"id":"W4312019575","doi":"10.2196/40102","title":"Comparison of Methods for Estimating Temporal Topic Models From Primary Care Clinical Text Data: Retrospective Closed Cohort Study","year":2022,"lang":"en","type":"article","venue":"JMIR Medical Informatics","topic":"Machine Learning in Healthcare","field":"Computer Science","cited_by":21,"is_retracted":false,"has_abstract":true,"ca_institutions":"Public Health Ontario; University of Toronto","funders":"Canadian Institutes of Health Research; Heart and Stroke Foundation of Canada","keywords":"Latent Dirichlet allocation; Topic model; Computer science; Artificial intelligence; Latent class model; Machine learning","routes":{"ca_aff":true,"ca_fund":true,"ca_venue":true,"about_ca":true,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":["metaresearch"],"consensus_categories":[],"category_scores_codex":[0.09162898,0.001223643,0.001406898,0.002525577,0.0009446641,0.002437775,0.002556261,0.001865476,0.001792511],"category_scores_gemma":[0.1985421,0.0009987868,0.003137946,0.001922001,0.0008743919,0.002222784,0.001873137,0.002059057,0.000456723],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001627424,"about_ca_system_score_gemma":0.002263389,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01909456,"about_ca_topic_score_gemma":0.0163763,"domain_scores_codex":[0.9751024,0.01911366,0.001327751,0.003255075,0.0009251675,0.0002759602],"domain_scores_gemma":[0.6589791,0.3159341,0.008158011,0.009318305,0.006285533,0.001325019],"domain_codex":null,"domain_gemma":"methods","domain_candidate":"methods","domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.0116324,0.001483403,0.6356579,0.001142686,0.007875631,0.0005090757,0.003275258,0.1177713,0.001387209,0.00387544,0.004470692,0.2109189],"study_design_scores_gemma":[0.001094984,0.001928022,0.1263469,0.0003800924,0.001977824,0.0006341395,0.001708636,0.8540792,0.00104404,0.00771008,0.00288692,0.0002091539],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.6331874,0.00263845,0.3567948,0.0007307748,0.0002228005,0.002202722,0.003100762,0.0003795537,0.0007426578],"genre_scores_gemma":[0.8066807,0.001265556,0.1806601,0.0003076239,0.0002191261,0.002426378,0.00750328,0.0001674605,0.0007697997],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.908371,"threshold_uncertainty_score":0.4845862,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1450194641126829,"score_gpt":0.5170725829912691,"score_spread":0.3720531188785862,"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."}}