{"id":"W4251929127","doi":"10.2196/preprints.22651","title":"Use of Automated Thematic Annotations for Small Data Sets in a Psychotherapeutic Context: Systematic Review of Machine Learning Algorithms (Preprint)","year":2020,"lang":"en","type":"preprint","venue":"","topic":"Mental Health via Writing","field":"Psychology","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Institut national de psychiatrie légale Philippe-Pinel; Université de Montréal; Institut Universitaire en Santé Mentale de Québec","funders":"","keywords":"Machine learning; Naive Bayes classifier; Preprint; PsycINFO; Computer science; Artificial intelligence; Context (archaeology); Information retrieval; Support vector machine; Systematic review; Algorithm; MEDLINE; Data science; World Wide Web","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":[0.0986002,0.001524944,0.007536879,0.03703951,0.001539551,0.005488486,0.00362108,0.001639647,0.005906218],"category_scores_gemma":[0.3448167,0.001316557,0.008903184,0.02024277,0.00255819,0.008144087,0.004502257,0.001744586,0.0007210315],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00739505,"about_ca_system_score_gemma":0.02356397,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004574177,"about_ca_topic_score_gemma":0.01540775,"domain_scores_codex":[0.9139846,0.04548379,0.02693995,0.003635915,0.009445203,0.00051052],"domain_scores_gemma":[0.513366,0.4152638,0.0421166,0.008341848,0.02022143,0.0006903284],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"systematic_review","study_design_gemma":"systematic_review","study_design_scores_codex":[0.0002006424,0.00003512175,0.001492579,0.8116953,0.005547694,0.00009304497,0.001472301,0.0003275727,0.0002276768,0.001004561,0.004055816,0.1738477],"study_design_scores_gemma":[0.0002265169,0.000172301,0.004053528,0.9443966,0.01847999,0.0002042883,0.001641611,0.0005937424,0.0005460295,0.002135316,0.02744885,0.0001013315],"study_design_candidate":"systematic_review","study_design_consensus":"systematic_review","genre_codex":"review","genre_gemma":"review","genre_scores_codex":[0.006827962,0.9588025,0.01320491,0.003649148,0.0008233588,0.01092054,0.003888525,0.0001786055,0.001704486],"genre_scores_gemma":[0.08670616,0.7952103,0.08035015,0.00314972,0.0005939346,0.02933952,0.003831948,0.000140449,0.0006778226],"genre_candidate":"review","genre_consensus":"review","teacher_disagreement_score":0.0986002,"threshold_uncertainty_score":0.521454,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.3280655431764375,"score_gpt":0.4583071346414419,"score_spread":0.1302415914650044,"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."}}