{"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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.002317868,0.0001509415,0.0003465821,0.0003320682,0.0002646277,0.00008190396,0.000132429,0.0001338473,0.0001687718],"category_scores_gemma":[0.001008999,0.0001384818,0.00007635143,0.0009328483,0.00009228589,0.00009084292,0.0002391425,0.001095146,0.000008674956],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001455584,"about_ca_system_score_gemma":0.0001973978,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001013838,"about_ca_topic_score_gemma":0.00004427262,"domain_scores_codex":[0.9971113,0.0004076239,0.0003755908,0.0004874695,0.001115774,0.0005022064],"domain_scores_gemma":[0.9985352,0.0001296628,0.00006319527,0.0003920522,0.0005831287,0.0002967511],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.00007137582,0.0003083433,0.968119,0.0001755881,0.00004773394,0.00001461769,0.002573955,0.00008609648,0.02568472,0.0001078925,0.0002058649,0.002604858],"study_design_scores_gemma":[0.002765857,0.000614468,0.9217638,0.0005428065,0.0002169272,0.0001123776,0.009249073,0.0007597219,0.0608304,0.00005173404,0.002772894,0.0003199237],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9934381,0.0005429403,0.0002841528,0.000412965,0.00009690467,0.0003985165,0.00002933975,0.00005517662,0.004741907],"genre_scores_gemma":[0.9975925,0.0001033026,0.0004068456,0.00006270337,0.000183115,0.00003552588,0.0002557751,0.00003837995,0.001321841],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.04635515,"threshold_uncertainty_score":0.5647126,"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."}}