{"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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0003795127,0.0001087824,0.0001666257,0.0001189522,0.0004071744,0.00007700537,0.00006089217,0.00009636533,0.00003489448],"category_scores_gemma":[0.00002258062,0.0001155024,0.00001508589,0.0001154096,0.00002996675,0.0002470046,0.00004164848,0.0001305636,0.000002141298],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000107791,"about_ca_system_score_gemma":0.0000150895,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0002653196,"about_ca_topic_score_gemma":0.00005687011,"domain_scores_codex":[0.9988527,0.0001674733,0.0002730202,0.0003960851,0.0001219886,0.0001887147],"domain_scores_gemma":[0.999413,0.0001121003,0.00009318894,0.000257405,0.00005465528,0.0000696752],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"bench_or_experimental","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.000228054,0.01867744,0.04382513,0.000330502,0.000156249,0.0001861869,0.005813476,4.580768e-8,0.4679678,0.0001666406,0.00004808953,0.4626004],"study_design_scores_gemma":[0.002454033,0.002254433,0.02611707,0.0004414103,0.0002006923,0.0002881541,0.1489595,0.004500072,0.8142268,0.0001393348,0.00006563944,0.0003528541],"study_design_candidate":"bench_or_experimental","study_design_consensus":"bench_or_experimental","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9922996,0.000310247,0.003944736,0.00004807057,0.0001158472,0.001686632,0.000003831793,0.000204324,0.00138675],"genre_scores_gemma":[0.9845648,0.000001553985,0.01368506,0.0001019187,0.0001368589,0.001333327,0.0000086481,0.00001996115,0.0001478973],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.4622475,"threshold_uncertainty_score":0.4710053,"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."}}