{"id":"W4388773811","doi":"10.1016/j.eswa.2023.122582","title":"A machine learning tool for collecting and analyzing subjective road safety data from Twitter","year":2023,"lang":"en","type":"article","venue":"Expert Systems with Applications","topic":"Sentiment Analysis and Opinion Mining","field":"Computer Science","cited_by":19,"is_retracted":false,"has_abstract":false,"ca_institutions":"University of Saskatchewan","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Computer science; Support vector machine; Naive Bayes classifier; Random forest; Artificial intelligence; Machine learning; Crowdsourcing; Classifier (UML); Social media; World Wide Web","routes":{"ca_aff":true,"ca_fund":true,"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":[],"consensus_categories":[],"category_scores_codex":[0.001670942,0.0009625203,0.0007740296,0.005708499,0.0008582306,0.001072842,0.0007969167,0.0008010768,0.004289878],"category_scores_gemma":[0.00632429,0.0003416968,0.0005893821,0.003741586,0.0002116992,0.001856743,0.001038287,0.0008091241,0.004484753],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0005990501,"about_ca_system_score_gemma":0.001051686,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003869843,"about_ca_topic_score_gemma":0.00900937,"domain_scores_codex":[0.9987803,0.0002123035,0.0001969787,0.0002161927,0.0004995401,0.00009465092],"domain_scores_gemma":[0.9960198,0.001924981,0.0004221922,0.0003502032,0.001097207,0.0001855589],"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":[0.0006458568,0.001325898,0.05346175,0.00102817,0.0003654498,0.0006359871,0.001150921,0.008451971,0.05721503,0.003022414,0.1128956,0.7598009],"study_design_scores_gemma":[0.0001968953,0.0007758784,0.08696429,0.0001921913,0.0003120564,0.0007060295,0.001522384,0.7566358,0.06002112,0.009328162,0.0831401,0.0002051053],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.2010177,0.0004621201,0.6156521,0.001260695,0.0004093519,0.003436061,0.09401327,0.07014266,0.01360596],"genre_scores_gemma":[0.2939183,0.0002785205,0.6374465,0.0003740795,0.0002667364,0.003135832,0.05457849,0.0005360654,0.009465471],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.005708499,"threshold_uncertainty_score":0.01435101,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.05363648456002865,"score_gpt":0.3167575060949625,"score_spread":0.2631210215349339,"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."}}