{"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":"metacan-v3-hybrid-931329e0061c","candidate_categories":["metaresearch"],"consensus_categories":[],"category_scores_codex":[0.1255769,0.001059339,0.001036875,0.003986501,0.001409173,0.004783149,0.002018123,0.001400148,0.002470795],"category_scores_gemma":[0.3746887,0.000794459,0.00161845,0.004095034,0.002156748,0.004622819,0.004571861,0.001597572,0.00189972],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001319134,"about_ca_system_score_gemma":0.00202261,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004675139,"about_ca_topic_score_gemma":0.006608935,"domain_scores_codex":[0.8562906,0.102979,0.01224701,0.01313627,0.01422097,0.001126115],"domain_scores_gemma":[0.4584698,0.3408371,0.05292489,0.07099705,0.07481352,0.001957554],"domain_codex":null,"domain_gemma":"methods","domain_candidate":"methods","domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.00185361,0.0004550887,0.7581204,0.001149547,0.001976505,0.0001520826,0.01507191,0.005964457,0.004058208,0.004067879,0.007635596,0.1994948],"study_design_scores_gemma":[0.0002365308,0.002119815,0.8164542,0.001142091,0.0008236077,0.0007829532,0.007773908,0.1183197,0.01391282,0.01308518,0.02483767,0.000511476],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.7226863,0.001537304,0.2477023,0.0008464744,0.0009013333,0.003254694,0.003803392,0.002107906,0.01716043],"genre_scores_gemma":[0.9156711,0.0002671898,0.07889476,0.0002791372,0.000172649,0.001230688,0.001888601,0.0002914036,0.001304569],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.8744231,"threshold_uncertainty_score":0.6641218,"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."}}