{"id":"W4399114217","doi":"10.1109/access.2024.3406469","title":"Detection and Prediction of Future Mental Disorder From Social Media Data Using Machine Learning, Ensemble Learning, and Large Language Models","year":2024,"lang":"en","type":"article","venue":"IEEE Access","topic":"Mental Health via Writing","field":"Psychology","cited_by":30,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Ottawa","funders":"","keywords":"Machine learning; Artificial intelligence; Ensemble learning; Computer science; Task (project management); Ensemble forecasting; Feeling; Social media; Mental illness; Support vector machine; Logistic regression; Mental health; Psychology; Natural language processing; Social psychology; World Wide Web; Psychiatry","routes":{"ca_aff":true,"ca_fund":false,"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":[0.0003208184,0.0001240644,0.0001667282,0.00009383953,0.0002449824,0.00007705197,0.0001343591,0.0001410836,0.0002196857],"category_scores_gemma":[0.00001629075,0.0001247092,0.00001792944,0.0001305697,0.00004009564,0.0005224631,0.0001844098,0.0004190145,0.000003578857],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00003360768,"about_ca_system_score_gemma":0.0000129586,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002709549,"about_ca_topic_score_gemma":0.001852959,"domain_scores_codex":[0.998799,0.0001812455,0.0002331599,0.0004042139,0.0001664302,0.0002159451],"domain_scores_gemma":[0.9995981,0.0001060563,0.00009966127,0.0001271283,0.00001386703,0.00005520455],"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":[0.000740903,0.0004402305,0.2265612,0.001488599,0.0005679228,0.0001200576,0.1565258,0.0001371344,0.08746403,0.0007073934,0.001356512,0.5238901],"study_design_scores_gemma":[0.003383621,0.0002088194,0.0308628,0.0003156836,0.0002885669,0.0001002666,0.02594666,0.9183869,0.003296578,0.0005891917,0.01610939,0.0005114694],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9865587,0.006954657,0.003875642,0.00005149698,0.001381211,0.0001716428,0.0005831127,0.0001064755,0.0003170086],"genre_scores_gemma":[0.998331,0.000160782,0.00008702226,0.00002393894,0.0008909248,0.000007879149,0.000384136,0.00003015368,0.00008413201],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.9182498,"threshold_uncertainty_score":0.5085493,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.06053425226155375,"score_gpt":0.394200777211908,"score_spread":0.3336665249503543,"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."}}