{"id":"W4213441134","doi":"10.2196/34492","title":"Traditional Machine Learning Models and Bidirectional Encoder Representations From Transformer (BERT)–Based Automatic Classification of Tweets About Eating Disorders: Algorithm Development and Validation Study","year":2022,"lang":"en","type":"article","venue":"JMIR Medical Informatics","topic":"Mental Health via Writing","field":"Psychology","cited_by":49,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Computer science; Encoder; Algorithm; Artificial intelligence; Machine learning; Natural language processing; Operating system","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":["insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.0008243421,0.0001372613,0.0002153589,0.0001863105,0.0005244203,0.00002470605,0.00009938156,0.00007415203,0.001238406],"category_scores_gemma":[0.00004511429,0.0001411816,0.00002788915,0.0002202419,0.00008604036,0.0002785051,0.00003741391,0.0004298785,0.000003965772],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00009703271,"about_ca_system_score_gemma":0.0001508664,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001876383,"about_ca_topic_score_gemma":0.00003450879,"domain_scores_codex":[0.9973837,0.0002636282,0.001033182,0.0001723604,0.0009516332,0.0001955331],"domain_scores_gemma":[0.9988008,0.0005526747,0.0003429296,0.00011114,0.00003826934,0.0001542328],"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":[0.00007484992,0.002083282,0.08785184,0.0004158798,0.0002198763,0.000004075568,0.1928127,0.001337043,0.00004262335,0.001092863,0.000178968,0.713886],"study_design_scores_gemma":[0.001975705,0.0002094403,0.1287972,0.0000792619,0.00002692087,0.00001423131,0.0378457,0.8301884,0.0000223283,0.0004755909,0.000213176,0.0001520486],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9567391,0.00008301312,0.04022531,0.0001892911,0.0001431683,0.0008454222,0.00008031266,0.00007215055,0.001622177],"genre_scores_gemma":[0.9810605,0.000006839196,0.01705468,0.0001558059,0.00003402855,0.0008041221,0.0008309953,0.00001522389,0.00003781591],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.8288513,"threshold_uncertainty_score":0.9996746,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.08479346342406621,"score_gpt":0.3703131037966668,"score_spread":0.2855196403726006,"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."}}