{"id":"W2759606429","doi":"10.1097/jom.0000000000001162","title":"Applying Machine Learning to Workers’ Compensation Data to Identify Industry-Specific Ergonomic and Safety Prevention Priorities","year":2017,"lang":"en","type":"article","venue":"Journal of Occupational and Environmental Medicine","topic":"Occupational Health and Safety Research","field":"Health Professions","cited_by":32,"is_retracted":false,"has_abstract":true,"ca_institutions":"Bell (Canada)","funders":"National Institutes of Health","keywords":"Causation; Occupational safety and health; Occupational injury; Workers' compensation; Human factors and ergonomics; Injury prevention; Psychological intervention; Compensation (psychology); Occupational medicine; Poison control; Environmental health; Medicine; Actuarial science; Business; Psychology; Nursing","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":"codex-gemma-dda1882f352a","candidate_categories":["sts"],"consensus_categories":[],"category_scores_codex":[0.001925429,0.0001234986,0.0002802801,0.0001759546,0.001526611,0.0000245092,0.000269217,0.0001229264,0.0004375199],"category_scores_gemma":[0.0004727308,0.000100118,0.00001882326,0.00004656398,0.0001273831,0.0004390608,0.0004039786,0.0008286607,0.00003023286],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001953631,"about_ca_system_score_gemma":0.0001088153,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001315122,"about_ca_topic_score_gemma":0.00005278813,"domain_scores_codex":[0.9980845,0.0002052583,0.0007179049,0.0002242578,0.0005310513,0.0002369856],"domain_scores_gemma":[0.9983215,0.0004579393,0.0005064702,0.0002569767,0.0000462712,0.0004108711],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.002034485,0.00003174011,0.9328434,0.00007028038,0.00002470637,0.000004779169,0.0007695131,0.00007088938,0.001085991,0.0000988523,0.001188167,0.06177722],"study_design_scores_gemma":[0.001420692,0.000364617,0.9537749,0.0005120119,0.00001573866,0.00001803296,0.00163105,0.0003618912,0.000002904305,0.00009117998,0.04172366,0.00008336035],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9912314,0.0005662988,0.001165968,0.005363323,0.0004648579,0.0008686048,0.00009226766,0.000005079964,0.0002422247],"genre_scores_gemma":[0.9957731,0.0008398388,0.001280618,0.0003929177,0.0008872753,0.00003207392,0.0001906878,0.00001223099,0.0005911983],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.06169386,"threshold_uncertainty_score":0.9997733,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.2359767467389541,"score_gpt":0.4937862900134714,"score_spread":0.2578095432745173,"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."}}