{"id":"W4389855122","doi":"10.1101/2023.12.15.23300047","title":"Using Natural Language Processing as a Scalable Mental Status Evaluation Technique","year":2023,"lang":"en","type":"preprint","venue":"medRxiv","topic":"Mental Health via Writing","field":"Psychology","cited_by":3,"is_retracted":false,"has_abstract":true,"ca_institutions":"Queen's University","funders":"","keywords":"Mental health; Scalability; Computer science; Natural language; Natural language processing; Psychology; Artificial intelligence; Machine learning; Applied psychology; Data science; 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":["metaepi_narrow","insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.002513026,0.0003644399,0.0003993411,0.0003411159,0.0002618894,0.0001010654,0.0003358618,0.0004248445,0.0009763906],"category_scores_gemma":[0.0002015811,0.000387504,0.0001056694,0.000353394,0.00007473821,0.0001027076,0.0005639721,0.001245946,0.0004951568],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0009689407,"about_ca_system_score_gemma":0.0004932748,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002721905,"about_ca_topic_score_gemma":0.0001122938,"domain_scores_codex":[0.9962015,0.0005809916,0.0006443183,0.0009152724,0.0008045677,0.0008533577],"domain_scores_gemma":[0.9985375,0.00008268636,0.0004650862,0.0005963669,0.000144062,0.0001742379],"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":[0.00084147,0.001179339,0.2022407,0.01046379,0.0006258022,0.001256071,0.09679557,0.000559476,0.1773595,0.0005410999,0.004586352,0.5035508],"study_design_scores_gemma":[0.01227907,0.0009069451,0.243471,0.02820271,0.001800476,0.00186208,0.09243112,0.5104508,0.0801917,0.01353555,0.005609094,0.009259438],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9793099,0.005215489,0.0003956584,0.0001748739,0.003405933,0.003012837,0.00005846873,0.000548374,0.007878455],"genre_scores_gemma":[0.9907455,0.00001347863,0.005387564,0.0002009296,0.0004306746,0.001022105,0.0003421624,0.0001410734,0.00171649],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.5098913,"threshold_uncertainty_score":0.9999369,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1325454369315427,"score_gpt":0.4818553390130314,"score_spread":0.3493099020814887,"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."}}