{"id":"W4383093199","doi":"10.20944/preprints202307.0192.v1","title":"Examining the Potential of Generative Language Models for Aviation Safety Analysis: Case Study and Insights using the Aviation Safety Reporting System (ASRS)","year":2023,"lang":"en","type":"preprint","venue":"Preprints.org","topic":"Topic Modeling","field":"Computer Science","cited_by":5,"is_retracted":false,"has_abstract":true,"ca_institutions":"General Fusion (Canada)","funders":"","keywords":"Aviation; Computer science; Aviation safety; Similarity (geometry); Crew; Process (computing); Aviation accident; Risk analysis (engineering); Aeronautics; Artificial intelligence; Engineering; Business","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":[0.005273785,0.0003114739,0.0006418278,0.0002705712,0.0006618861,0.0001387854,0.000811486,0.0001777271,0.000001868436],"category_scores_gemma":[0.0004161524,0.0002208349,0.0002270013,0.0005268526,0.00004502255,0.0003181925,0.002637151,0.0004611488,0.000001402423],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00025043,"about_ca_system_score_gemma":0.0001703478,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002566561,"about_ca_topic_score_gemma":0.0003987851,"domain_scores_codex":[0.995246,0.0007521029,0.001943494,0.001233841,0.0005603397,0.0002642106],"domain_scores_gemma":[0.9941561,0.0003504175,0.003008362,0.002034106,0.0003955849,0.00005545763],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00001972445,0.00004219625,0.02050419,0.0001532139,0.0008664218,0.000148006,0.07186291,0.9020393,0.001440306,0.001891588,1.991828e-7,0.00103198],"study_design_scores_gemma":[0.0002545738,0.00001328856,0.01783803,0.000105555,0.0006970798,0.00008421407,0.02400608,0.9551496,0.0006596576,0.0009843947,4.024326e-7,0.0002071448],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.5038524,0.00005868349,0.4945626,0.00004562232,0.0002688738,0.001085395,0.000008945885,0.0000917131,0.00002577193],"genre_scores_gemma":[0.9913384,0.00001004366,0.008202116,0.0000152125,0.0001670535,0.0001665531,0.00002356522,0.00002628386,0.00005071511],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.487486,"threshold_uncertainty_score":0.9005384,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.2380561019543551,"score_gpt":0.3699562733530729,"score_spread":0.1319001713987177,"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."}}