{"id":"W4214853627","doi":"10.1109/ieeeconf53345.2021.9723273","title":"Improving Depression Assessment With Multi-Task Learning From Speech and Text Information","year":2021,"lang":"en","type":"article","venue":"2021 55th Asilomar Conference on Signals, Systems, and Computers","topic":"Mental Health via Writing","field":"Psychology","cited_by":8,"is_retracted":false,"has_abstract":true,"ca_institutions":"Queen's University","funders":"","keywords":"Depression (economics); Task (project management); Computer science; Machine learning; Artificial intelligence; Complement (music); Distress; Feature engineering; Intervention (counseling); Feature (linguistics); Baseline (sea); Calibration; Natural language processing; Deep learning; Psychology; Clinical psychology; Psychiatry; Statistics","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":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0003977029,0.0002747915,0.0004210444,0.0001160851,0.0003156416,0.0004147399,0.0001159712,0.000143648,0.0001637456],"category_scores_gemma":[0.00002051166,0.0002477955,0.0000294226,0.0001327697,0.00005494801,0.0003267929,0.0001290918,0.0004075677,0.00005676257],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00007400481,"about_ca_system_score_gemma":0.0001545861,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001505109,"about_ca_topic_score_gemma":0.00005170411,"domain_scores_codex":[0.9976292,0.0005479261,0.0005424314,0.0005442723,0.0003525981,0.0003836277],"domain_scores_gemma":[0.9986618,0.0002512789,0.0003999596,0.0002772716,0.0001736393,0.0002360648],"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":[0.000292156,0.0002547096,0.1183765,0.0008308272,0.0002737301,0.0003984622,0.005171007,0.0003868777,0.004190191,0.00380683,0.0005815062,0.8654372],"study_design_scores_gemma":[0.01301862,0.002022074,0.2504988,0.007436275,0.0002128752,0.0006894282,0.102079,0.6137459,0.001285023,0.00007345468,0.006822889,0.002115705],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.8219758,0.002365021,0.1659305,0.0002186625,0.001217768,0.0009689667,0.00005429929,0.0001351278,0.007133804],"genre_scores_gemma":[0.9929186,0.00006064475,0.006191893,0.0002825051,0.0001338225,0.00005816653,0.0001505305,0.00002052508,0.0001833121],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.8633215,"threshold_uncertainty_score":0.9999974,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04026029015149168,"score_gpt":0.3161202145563112,"score_spread":0.2758599244048195,"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."}}