{"id":"W4396767426","doi":"10.1016/j.autcon.2024.105458","title":"Data-driven automatic classification model for construction accident cases using natural language processing with hyperparameter tuning","year":2024,"lang":"en","type":"article","venue":"Automation in Construction","topic":"Occupational Health and Safety Research","field":"Health Professions","cited_by":29,"is_retracted":false,"has_abstract":false,"ca_institutions":"University of Toronto","funders":"Korea Institute for Advancement of Technology; Ministry of Trade, Industry and Energy","keywords":"Hyperparameter; Accident (philosophy); Computer science; Artificial intelligence; Machine learning; Natural language processing; Data mining","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":[0.0008194814,0.0001790655,0.0002400338,0.0005200344,0.0007020286,0.00009569574,0.0001676355,0.0001825796,0.00005225066],"category_scores_gemma":[0.0005309019,0.0001569289,0.00002943565,0.0006143714,0.0001298219,0.001460116,0.00006097136,0.0005109237,0.00002479738],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0006665891,"about_ca_system_score_gemma":0.001531985,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001550203,"about_ca_topic_score_gemma":0.0003784094,"domain_scores_codex":[0.9976265,0.0002625211,0.0007681914,0.0005130873,0.0004025909,0.0004271328],"domain_scores_gemma":[0.9977807,0.001197866,0.000284554,0.0003495754,0.0003000644,0.00008728652],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0002717258,0.00004078468,0.03985942,0.003284946,0.0000446639,0.00001228715,0.003647637,0.009765177,0.002962635,0.003864172,0.0002419723,0.9360046],"study_design_scores_gemma":[0.0005697011,0.00003824961,0.01698417,0.001029909,0.00003978787,0.0001349932,0.003283117,0.9773385,0.00002270581,0.000297875,0.00009448165,0.0001664907],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.7702603,0.0003321352,0.2250728,0.0006443707,0.0009393306,0.002014963,0.0001298718,0.0004445572,0.0001616754],"genre_scores_gemma":[0.8590793,0.00001330114,0.139512,0.0000749292,0.0002120772,0.0003452688,0.0006827569,0.00003252397,0.00004788006],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.9675733,"threshold_uncertainty_score":0.6399373,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.2022277656843931,"score_gpt":0.4993653301654243,"score_spread":0.2971375644810312,"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."}}