{"id":"W1976129119","doi":"10.1002/ajim.22005","title":"Performance of automated and manual coding systems for occupational data: A case study of historical records","year":2011,"lang":"en","type":"article","venue":"American Journal of Industrial Medicine","topic":"Health, Environment, Cognitive Aging","field":"Environmental Science","cited_by":30,"is_retracted":false,"has_abstract":true,"ca_institutions":"McGill University","funders":"National Heart, Lung, and Blood Institute; National Institute for Occupational Safety and Health; U.S. Public Health Service; National Institutes of Health","keywords":"Coding (social sciences); Medicine; Census; Software; Environmental health; Computer science; Statistics; Population; Mathematics; Programming language","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":[0.08945867,0.0004916267,0.0005443806,0.004316659,0.002504535,0.002610549,0.002085858,0.001295767,0.0008945253],"category_scores_gemma":[0.3118685,0.0007210009,0.0007346573,0.003582741,0.00205632,0.002054971,0.002633463,0.001194539,0.0003267216],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.003278207,"about_ca_system_score_gemma":0.002601991,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01906619,"about_ca_topic_score_gemma":0.01883169,"domain_scores_codex":[0.8843275,0.08844665,0.007207091,0.004225991,0.01383356,0.001959289],"domain_scores_gemma":[0.5994204,0.2812729,0.04083426,0.03485373,0.04148606,0.002132605],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.0006207861,0.0005237104,0.8734254,0.000303346,0.0001743009,0.001470036,0.0363642,0.00101629,0.0007881809,0.0007699028,0.001595529,0.08294836],"study_design_scores_gemma":[0.0002022692,0.002787536,0.9022983,0.0009876507,0.0004077445,0.006456034,0.04127931,0.02919713,0.006340452,0.002214462,0.007518438,0.0003107354],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9931678,0.0003531123,0.004539568,0.0002559131,0.00002912196,0.0002897868,0.0001708435,0.00003831705,0.001155549],"genre_scores_gemma":[0.9881415,0.0001914494,0.01080506,0.0001062874,0.00002758881,0.0002475397,0.0002240581,0.00003634884,0.0002202014],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.08945867,"threshold_uncertainty_score":0.4731084,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.169129915408648,"score_gpt":0.3482634161337277,"score_spread":0.1791335007250797,"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."}}