{"id":"W4410808512","doi":"10.1371/journal.pone.0322299","title":"Assessing ML classification algorithms and NLP techniques for depression detection: An experimental case study","year":2025,"lang":"en","type":"article","venue":"PLoS ONE","topic":"Mental Health via Writing","field":"Psychology","cited_by":9,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"","keywords":"Machine learning; Random forest; Support vector machine; Feature selection; Context (archaeology); Artificial intelligence; Computer science; Depression (economics); Mental health; Anxiety; Distress; Statistical classification; Economic shortage; Feature (linguistics); Feature engineering; Natural language processing; Medicine; Psychiatry; Clinical psychology; Deep learning","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.008205567,0.001114797,0.0006486415,0.001186717,0.0008678702,0.001340776,0.001719754,0.002011691,0.001729488],"category_scores_gemma":[0.01975876,0.0003533049,0.001218541,0.0009675421,0.001318668,0.001254614,0.001322907,0.001436553,0.0009545767],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001532851,"about_ca_system_score_gemma":0.0007045184,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.005878482,"about_ca_topic_score_gemma":0.007295032,"domain_scores_codex":[0.9931521,0.003721842,0.0006837453,0.0009695828,0.001162727,0.0003099738],"domain_scores_gemma":[0.9757121,0.0156005,0.001129529,0.002845561,0.004331618,0.000380646],"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":[0.009321949,0.02814197,0.1585354,0.006000697,0.001486274,0.00486671,0.007134701,0.1875103,0.04363463,0.006553573,0.03329285,0.513521],"study_design_scores_gemma":[0.003201881,0.02817169,0.1182137,0.0008030094,0.001243425,0.003975995,0.01136288,0.6478208,0.1280822,0.007854479,0.04886029,0.000409738],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9633291,0.0006435408,0.0283413,0.000554615,0.0001560461,0.00172632,0.001931046,0.0004687846,0.002849275],"genre_scores_gemma":[0.887519,0.0004470682,0.1022701,0.0002951264,0.0001166966,0.001749008,0.004894149,0.00009833442,0.002610391],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.008205567,"threshold_uncertainty_score":0.0433957,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.2109665596265453,"score_gpt":0.4696552283861907,"score_spread":0.2586886687596454,"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."}}