{"id":"W4319057825","doi":"10.1093/annweh/wxad002","title":"Automated Coding of Job Descriptions From a General Population Study: Overview of Existing Tools, Their Application and Comparison","year":2023,"lang":"en","type":"article","venue":"Annals of Work Exposures and Health","topic":"Reliability and Agreement in Measurement","field":"Decision Sciences","cited_by":13,"is_retracted":false,"has_abstract":true,"ca_institutions":"Saint John Regional Hospital; University of Toronto; University of New Brunswick","funders":"National Cancer Institute; National Institutes of Health; European Commission; University of Warwick","keywords":"Coding (social sciences); Computer science; Population; Statistics; Medicine; Mathematics; Environmental health","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":[],"consensus_categories":[],"category_scores_codex":[0.004612302,0.00008518875,0.0004461333,0.0001270009,0.0001521914,0.00004162795,0.0001377939,0.00003576022,0.000007713687],"category_scores_gemma":[0.0002590264,0.0000637132,0.00004763166,0.000608271,0.00004349641,0.000108846,0.00007120289,0.00005507213,9.036414e-7],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000008376076,"about_ca_system_score_gemma":0.00002682828,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002259393,"about_ca_topic_score_gemma":0.0004359836,"domain_scores_codex":[0.9976774,0.0002782391,0.001085185,0.0002462947,0.0005727698,0.0001400801],"domain_scores_gemma":[0.9983966,0.0003815823,0.0006669066,0.000261794,0.0002311596,0.00006195669],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.00002709264,0.0001218334,0.9161073,0.00007438548,0.00002179034,5.599039e-8,0.002414081,0.0004472137,0.0003449634,0.001076703,0.001961903,0.07740267],"study_design_scores_gemma":[0.0001836033,0.0002389209,0.9864707,0.0001840888,0.000006496465,6.928525e-8,0.002428269,0.005012978,0.000106004,0.005071063,0.0002473901,0.00005039408],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9950069,0.002613239,0.000395315,0.001402352,0.00005270962,0.0004181844,0.00003929924,0.00003095527,0.00004099845],"genre_scores_gemma":[0.9982296,0.00098059,0.000635661,0.00008017892,0.00002489838,0.00001066637,0.00002328996,0.000003807472,0.00001129549],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.07735227,"threshold_uncertainty_score":0.3415542,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.7392532214113784,"score_gpt":0.5341445500158267,"score_spread":0.2051086713955517,"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."}}