{"id":"W2909965413","doi":"10.1139/cjce-2018-0117","title":"An extreme gradient boosting method for identifying the factors contributing to crash/near-crash events: a naturalistic driving study","year":2019,"lang":"en","type":"article","venue":"Canadian Journal of Civil Engineering","topic":"Traffic and Road Safety","field":"Engineering","cited_by":58,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Multicollinearity; Crash; Machine learning; Boosting (machine learning); Computer science; Classifier (UML); Artificial intelligence; Statistics; Mathematics; Regression analysis","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":true,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.001363052,0.0003020806,0.0004413231,0.0003221325,0.0003296798,0.0001913176,0.0005177523,0.00008492146,0.00004265334],"category_scores_gemma":[0.0005029434,0.0002484626,0.0001828435,0.0003432322,0.00001004217,0.0003256677,0.0000240245,0.0005307375,0.000004333521],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0004535637,"about_ca_system_score_gemma":0.000165337,"about_ca_topic_candidate":true,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0003537854,"about_ca_topic_score_gemma":0.02999865,"domain_scores_codex":[0.997945,0.00005317473,0.0006340937,0.0002071341,0.0002598626,0.0009007059],"domain_scores_gemma":[0.9982692,0.000443502,0.0001251996,0.0002794311,0.0001778721,0.0007047706],"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":[0.000003691621,0.00001138015,0.06240103,0.00008135336,0.0001966925,0.00002978232,0.004170051,0.9305125,0.001623614,0.00007998012,0.0001115336,0.00077843],"study_design_scores_gemma":[0.001343126,0.0003201426,0.4832126,0.0008047887,0.0002119881,0.0001291915,0.005602125,0.5034851,0.0002628411,0.00004000053,0.003767808,0.0008202868],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.7469289,0.0003399414,0.2497806,0.00004453342,0.002217023,0.0005481672,0.00001838616,0.00007814496,0.00004431209],"genre_scores_gemma":[0.9962928,0.00000226686,0.003323633,0.00002357695,0.0002354789,0.00001020141,0.000004193984,0.00008720588,0.00002067589],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.4270274,"threshold_uncertainty_score":0.9999968,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01905352178618235,"score_gpt":0.2513812300012569,"score_spread":0.2323277082150745,"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."}}