{"id":"W4387815716","doi":"10.1177/21695067231193645","title":"Improving Patient Safety Event Report Classification with Machine Learning and Contextual Text Representation","year":2023,"lang":"en","type":"article","venue":"Proceedings of the Human Factors and Ergonomics Society Annual Meeting","topic":"Patient Safety and Medication Errors","field":"Health Professions","cited_by":4,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"Agency for Healthcare Research and Quality","keywords":"Artificial intelligence; Classifier (UML); Computer science; Machine learning; Support vector machine; Confusion; Confusion matrix; Natural language processing; F1 score; Representation (politics); Psychology","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":[],"consensus_categories":[],"category_scores_codex":[0.001470981,0.000802079,0.000631399,0.002639787,0.0003710232,0.001235155,0.0007135937,0.0007956526,0.001010858],"category_scores_gemma":[0.01206604,0.0001515859,0.0007340284,0.001879927,0.0002634542,0.001959698,0.0006850472,0.0008418495,0.0006903299],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0005344928,"about_ca_system_score_gemma":0.0009417156,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003037068,"about_ca_topic_score_gemma":0.002683995,"domain_scores_codex":[0.9986392,0.0005182173,0.0001725589,0.0002920871,0.000276003,0.0001019762],"domain_scores_gemma":[0.9942937,0.003270079,0.0006224507,0.0005243168,0.001193561,0.00009589214],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0003132948,0.0005395733,0.012377,0.0002030425,0.0000866618,0.0002208391,0.0002426484,0.03968577,0.01228868,0.001048786,0.00515601,0.9278377],"study_design_scores_gemma":[0.00003269579,0.0003523943,0.00948063,0.00005061107,0.0001167459,0.0001426094,0.0002588929,0.9654776,0.01667492,0.003830327,0.003538153,0.00004451252],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.4082909,0.00236694,0.5707768,0.002008572,0.0004912509,0.0004832766,0.001731421,0.009292972,0.004557775],"genre_scores_gemma":[0.8026,0.0005637748,0.1921491,0.0002054099,0.0003307138,0.0001922964,0.00276316,0.00006141856,0.001134041],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.003037068,"threshold_uncertainty_score":0.007779419,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04040348190884215,"score_gpt":0.3310598994626997,"score_spread":0.2906564175538576,"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."}}