{"id":"W4285364837","doi":"10.2196/33678","title":"Classification of Twitter Vaping Discourse Using BERTweet: Comparative Deep Learning Study","year":2022,"lang":"en","type":"article","venue":"JMIR Medical Informatics","topic":"Data-Driven Disease Surveillance","field":"Medicine","cited_by":20,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"U.S. National Library of Medicine; National Cancer Institute; National Institutes of Health; National Science Foundation","keywords":"Computer science; Artificial intelligence; Deep learning; Machine learning; Relevance (law); Classifier (UML); Sentiment analysis; Random forest; Natural language processing; Language model; Transfer of learning; Transformer; Information retrieval","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":[],"consensus_categories":[],"category_scores_codex":[0.0006865872,0.000146194,0.0004581835,0.0001567788,0.0001892816,0.00001465713,0.000222657,0.00004807306,0.0008353168],"category_scores_gemma":[0.0001546252,0.0001273798,0.00008046134,0.0003899271,0.0001725341,0.0001905724,0.0002840596,0.0006388982,0.00002641016],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001464646,"about_ca_system_score_gemma":0.0002145092,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001500033,"about_ca_topic_score_gemma":0.000005557388,"domain_scores_codex":[0.9971959,0.000190405,0.0008340542,0.0001205265,0.001420442,0.0002386975],"domain_scores_gemma":[0.9987348,0.0001298286,0.000432182,0.0003432776,0.00011196,0.0002479884],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0007911134,0.005002616,0.6797223,0.00107996,0.0008250483,0.0001352378,0.27878,0.002862127,0.0006709791,0.0002828553,0.004764692,0.02508309],"study_design_scores_gemma":[0.003907793,0.0009323188,0.2145507,0.0002032525,0.0001871168,0.00007924483,0.2825486,0.4905889,0.00003180655,0.00001598275,0.006641042,0.0003132205],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9951563,0.00004053007,0.001871501,0.0001514519,0.0001132156,0.0007886041,0.00001838692,0.00007275095,0.001787269],"genre_scores_gemma":[0.9983697,0.000003293187,0.0007841977,0.0003806178,0.00006767113,0.00008857069,0.0002232816,0.0000144221,0.0000682624],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.4877268,"threshold_uncertainty_score":0.9146133,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.07220711212445817,"score_gpt":0.3981763846864982,"score_spread":0.32596927256204,"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."}}