{"id":"W4410185915","doi":"10.2196/63272","title":"Improving Suicidal Ideation Detection in Social Media Posts: Topic Modeling and Synthetic Data Augmentation Approach","year":2025,"lang":"en","type":"article","venue":"JMIR Formative Research","topic":"Mental Health via Writing","field":"Psychology","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"National Research Council Canada; University of Ottawa","funders":"Natural Sciences and Engineering Research Council of Canada; University of Ottawa","keywords":"Preprint; Ideation; Social media; Suicidal ideation; Psychology; Data science; Sociology; Computer science; Suicide prevention; Poison control; World Wide Web; Medicine; Cognitive science; Medical emergency","routes":{"ca_aff":true,"ca_fund":true,"ca_venue":true,"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":[0.006231917,0.001353418,0.001041102,0.002348315,0.0007144593,0.00161432,0.001757343,0.001742993,0.001177679],"category_scores_gemma":[0.01884388,0.0005091461,0.001989119,0.001362296,0.0007307763,0.001615717,0.001656736,0.002622532,0.0007826918],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0007943385,"about_ca_system_score_gemma":0.001027344,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004586486,"about_ca_topic_score_gemma":0.005384279,"domain_scores_codex":[0.9976004,0.001510021,0.0001465506,0.0004569998,0.0001584064,0.0001276425],"domain_scores_gemma":[0.977088,0.0198523,0.0006493572,0.0009835166,0.001157634,0.0002692202],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.001535996,0.00153468,0.07187507,0.001167426,0.0006253865,0.0006651535,0.004161734,0.4545377,0.008870338,0.009289073,0.01803064,0.4277068],"study_design_scores_gemma":[0.00002602172,0.00008811661,0.001644708,0.00003980141,0.00004521649,0.0000478688,0.0002456795,0.9913182,0.001201414,0.00372497,0.001598127,0.00001992025],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.4304598,0.002820046,0.5504199,0.003386436,0.000739628,0.000883735,0.005302054,0.003250641,0.002737632],"genre_scores_gemma":[0.8054373,0.0006078315,0.1799709,0.0006091793,0.0004710413,0.0009592961,0.01003891,0.0001427073,0.001762986],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.006231917,"threshold_uncertainty_score":0.03295791,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1686070679126437,"score_gpt":0.4998295271952565,"score_spread":0.3312224592826128,"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."}}