{"id":"W4200373041","doi":"10.1093/ntr/ntab257","title":"A Machine Learning Approach to Identify Predictors of Frequent Vaping and Vulnerable Californian Youth Subgroups","year":2021,"lang":"en","type":"article","venue":"Nicotine & Tobacco Research","topic":"Smoking Behavior and Cessation","field":"Medicine","cited_by":18,"is_retracted":false,"has_abstract":true,"ca_institutions":"Public Health Ontario; University of Toronto; Centre for Addiction and Mental Health; Sunnybrook Health Science Centre","funders":"National Institute on Drug Abuse; Canadian Institutes of Health Research; National Institutes of Health; National Cancer Institute; Tobacco-Related Disease Research Program","keywords":"Logistic regression; Demography; Nicotine; Medicine; Cohort; Random forest; Pacific islanders; Ethnic group; Psychology; Environmental health; Gerontology; Psychiatry; Internal medicine; Population","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":[0.003743613,0.000800336,0.0007962707,0.002601552,0.0006338578,0.0009076003,0.0007793382,0.0007682406,0.002180385],"category_scores_gemma":[0.005869797,0.0002788379,0.001544126,0.001102643,0.0002830089,0.0004681839,0.0005595088,0.000854289,0.0004099872],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0005631681,"about_ca_system_score_gemma":0.001598494,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01564073,"about_ca_topic_score_gemma":0.01205594,"domain_scores_codex":[0.9992529,0.0003361612,0.00005312298,0.000186416,0.00006936541,0.0001020944],"domain_scores_gemma":[0.9971676,0.001884622,0.0003157663,0.0001368003,0.0003508664,0.0001442954],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.0003761208,0.0005821877,0.849827,0.0001547723,0.0006910167,0.0001991594,0.0002357735,0.04312081,0.0006319127,0.0006551843,0.003552937,0.09997317],"study_design_scores_gemma":[0.0001465183,0.000654102,0.1967997,0.0001950698,0.0003899143,0.0003280099,0.0003964852,0.7931849,0.0006420029,0.005292969,0.001911861,0.00005849191],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.8518791,0.00110615,0.1374869,0.001457197,0.00009780924,0.0005525956,0.004453445,0.000841988,0.002124862],"genre_scores_gemma":[0.9535776,0.0002142544,0.04249451,0.0001287356,0.00007901304,0.0003767489,0.002424622,0.00002159188,0.0006829159],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.01564073,"threshold_uncertainty_score":0.03109938,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0981454790116647,"score_gpt":0.3786602871954755,"score_spread":0.2805148081838107,"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."}}