{"id":"W4288287756","doi":"10.48550/arxiv.1907.02581","title":"Transfer Learning for Risk Classification of Social Media Posts: Model\\n Evaluation Study","year":2019,"lang":"","type":"preprint","venue":"arXiv (Cornell University)","topic":"Mental Health via Writing","field":"Psychology","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Natural Sciences and Engineering Research Council of Canada; Centre for Addiction and Mental Health Foundation; Nvidia","keywords":"Computer science; Machine learning; Artificial intelligence; Social media; Triage; Benchmark (surveying); Task (project management); F1 score; Natural language processing; Transfer of learning; Sentence; World Wide Web; Psychology","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.006342993,0.001972541,0.001341534,0.001245072,0.0007376999,0.001206371,0.001756077,0.002058104,0.004679872],"category_scores_gemma":[0.01027969,0.0002906539,0.001065873,0.0007835153,0.0005535228,0.001771915,0.001665853,0.003153843,0.002628957],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001637579,"about_ca_system_score_gemma":0.001419667,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01615765,"about_ca_topic_score_gemma":0.01208447,"domain_scores_codex":[0.9984757,0.0006954888,0.00008154178,0.0003210788,0.00023445,0.0001916575],"domain_scores_gemma":[0.9945526,0.003494115,0.0001800647,0.0004633035,0.001082045,0.0002278412],"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":[0.003194093,0.002737798,0.02558143,0.0006942246,0.0008042423,0.0004157471,0.0003983426,0.3246697,0.003571252,0.002115319,0.04522697,0.5905908],"study_design_scores_gemma":[0.00006117443,0.000279585,0.001711402,0.00002724236,0.00004626584,0.0000295082,0.00007526543,0.994597,0.001316757,0.0011192,0.0007201304,0.00001644938],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.8192904,0.009147979,0.1362244,0.003948906,0.001402645,0.001152169,0.005156219,0.008641068,0.01503633],"genre_scores_gemma":[0.9413635,0.000736632,0.04006728,0.0005383677,0.000324997,0.0004589377,0.005902705,0.0002317734,0.01037585],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.01615765,"threshold_uncertainty_score":0.03354537,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.2598921337851723,"score_gpt":0.3199921065789157,"score_spread":0.06009997279374346,"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."}}