{"id":"W4401995522","doi":"10.1080/23249935.2024.2396951","title":"Predicting metro incident duration using structured data and unstructured text logs","year":2024,"lang":"en","type":"article","venue":"Transportmetrica A Transport Science","topic":"Traffic Prediction and Management Techniques","field":"Engineering","cited_by":3,"is_retracted":false,"has_abstract":true,"ca_institutions":"McGill University","funders":"Natural Science Basic Research Program of Shaanxi Province; National Natural Science Foundation of China","keywords":"Unstructured data; Duration (music); Computer science; Data mining; Big data; Physics","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":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.000583311,0.0002753987,0.000247152,0.001271795,0.0002173378,0.0001600357,0.0007077887,0.0001005075,0.00002718718],"category_scores_gemma":[0.00001993648,0.0002571006,0.00005236208,0.003748185,0.0003351862,0.001552911,0.00002958679,0.0002957775,0.00000183028],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001273267,"about_ca_system_score_gemma":0.0001025483,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00006234136,"about_ca_topic_score_gemma":0.0001252678,"domain_scores_codex":[0.9975652,0.000009592643,0.0005322935,0.0007513339,0.0007047775,0.0004368043],"domain_scores_gemma":[0.9991122,0.00003260838,0.00004083672,0.0005663648,0.00004653407,0.0002013892],"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":[0.0001181886,0.0001485425,0.2654779,0.003396745,0.0007093261,0.001146388,0.007330532,0.03819463,0.2881444,0.0125452,0.001160597,0.3816276],"study_design_scores_gemma":[0.0003047477,0.00004578473,0.2067252,0.0001030069,0.0002600137,0.00006919426,0.0001642451,0.7837158,0.004540483,0.0001435569,0.003468645,0.0004592712],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.6647649,0.001285068,0.3267581,0.00005114689,0.0008070527,0.0003999524,0.0002122978,0.005072013,0.0006494397],"genre_scores_gemma":[0.9893464,0.0003793323,0.01001163,0.00002254099,0.0000731239,0.000009303356,0.0001095819,0.0000396234,0.000008405786],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.7455212,"threshold_uncertainty_score":0.9999881,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02194726248391593,"score_gpt":0.2602541607326628,"score_spread":0.2383068982487468,"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."}}