{"id":"W2160078924","doi":"10.1093/annhyg/meu006","title":"Beyond Crosswalks: Reliability of Exposure Assessment Following Automated Coding of Free-Text Job Descriptions for Occupational Epidemiology","year":2014,"lang":"en","type":"article","venue":"The Annals of Occupational Hygiene","topic":"Occupational Health and Safety Research","field":"Health Professions","cited_by":35,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of British Columbia","funders":"","keywords":"Coding (social sciences); Computer science; Exposure assessment; Occupational safety and health; Statistics; Applied psychology; Environmental health; Medicine; Psychology; Mathematics","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":["metaresearch"],"consensus_categories":[],"category_scores_codex":[0.01167828,0.0002252517,0.0008360107,0.0002683356,0.0009159734,0.000004022908,0.0006759894,0.0002777595,0.0002099682],"category_scores_gemma":[0.01468205,0.000172317,0.0004067016,0.0005152373,0.000407843,0.0002474303,0.0002269766,0.0004200909,0.0000127472],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001034828,"about_ca_system_score_gemma":0.001772204,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0005123493,"about_ca_topic_score_gemma":0.0001028093,"domain_scores_codex":[0.9938161,0.001922135,0.002357211,0.0003908708,0.0008186735,0.0006949786],"domain_scores_gemma":[0.975011,0.01946465,0.001345698,0.0008719193,0.003112316,0.0001944488],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.002142741,0.0002959401,0.8774884,0.0009996376,0.00009713262,1.355955e-7,0.0003801693,0.004816098,0.0007513455,0.1025685,0.008808499,0.001651338],"study_design_scores_gemma":[0.001391892,0.0005087682,0.9336569,0.0002358716,0.00003241959,4.828312e-7,0.0001607796,0.03530374,0.0002547587,0.0272653,0.001058846,0.0001302389],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9654185,0.000200169,0.02197004,0.004989341,0.0006976047,0.001944781,0.001614445,0.00007506617,0.003090007],"genre_scores_gemma":[0.9851477,0.00004731573,0.01265217,0.0007502516,0.0002430036,0.0004543142,0.0004868284,0.00002434425,0.000194083],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.07530323,"threshold_uncertainty_score":0.9936177,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.2739037899492713,"score_gpt":0.5478444846014228,"score_spread":0.2739406946521515,"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."}}