{"id":"W2967730738","doi":"10.1037/pas0000764","title":"Scoring algorithms for a computer-based cognitive screening tool: An illustrative example of overfitting machine learning approaches and the impact on estimates of classification accuracy.","year":2019,"lang":"en","type":"article","venue":"Psychological Assessment","topic":"Dementia and Cognitive Impairment Research","field":"Medicine","cited_by":19,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto; University of Saskatchewan","funders":"Canadian Institutes of Health Research; University of Toronto; Consortium canadien en neurodégénérescence associée au vieillissement","keywords":"Overfitting; Logistic regression; Machine learning; Decision tree; Artificial intelligence; Medical diagnosis; Cross-validation; Regression; Psychology; Statistics; Computer science; Mathematics; Medicine; Artificial neural network","routes":{"ca_aff":true,"ca_fund":true,"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":[0.001618385,0.0001754632,0.0004129945,0.00008620622,0.000095864,0.00003787574,0.00008876454,0.00006119861,0.00007104166],"category_scores_gemma":[0.0002376347,0.00009683827,0.0001257228,0.0001487911,0.000223747,0.00009543758,0.00004359182,0.0002969021,8.229161e-7],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00002787739,"about_ca_system_score_gemma":0.00003365407,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00005236122,"about_ca_topic_score_gemma":3.622157e-7,"domain_scores_codex":[0.9983734,0.0002932144,0.0003456316,0.000400663,0.0003453622,0.0002416968],"domain_scores_gemma":[0.9966882,0.002598948,0.0002877185,0.0001549646,0.0001987512,0.00007140966],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.005474989,0.002120728,0.6777611,0.0002293579,0.0003464411,0.000003184534,0.0003305913,0.0008881568,0.005786767,0.001805617,0.00000398958,0.3052491],"study_design_scores_gemma":[0.005033093,0.005214003,0.5730759,0.0001948699,0.00005816205,0.000003701505,0.0007112374,0.414066,0.00145295,0.0001161932,0.000002278806,0.00007164737],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.7655244,0.00002583574,0.2319986,0.0001529077,0.00001682095,0.001427843,0.0000200065,0.00001944369,0.0008142038],"genre_scores_gemma":[0.9672302,0.000009405644,0.03236492,0.0000620349,0.00003895636,0.0001334635,0.0001371435,0.00001236965,0.00001152577],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.4131778,"threshold_uncertainty_score":0.394895,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.2068108851245482,"score_gpt":0.445626754101562,"score_spread":0.2388158689770138,"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."}}