{"id":"W4392108757","doi":"10.3389/fpubh.2024.1328353","title":"A feature optimization study based on a diabetes risk questionnaire","year":2024,"lang":"en","type":"article","venue":"Frontiers in Public Health","topic":"Artificial Intelligence in Healthcare","field":"Health Professions","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of British Columbia","funders":"Centers for Disease Control and Prevention; National Natural Science Foundation of China","keywords":"Medicine; Diabetes mellitus; Disease; Univariate; Feature selection; Standardization; Ranking (information retrieval); Risk factor; Computer science; Machine learning; Data mining; Environmental health; Internal medicine; Multivariate statistics","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"],"consensus_categories":[],"category_scores_codex":[0.005224926,0.0002814721,0.0004879834,0.001034987,0.001067675,0.00008131137,0.0003122274,0.0003633532,0.000134089],"category_scores_gemma":[0.001993146,0.0002539153,0.00007648888,0.001790088,0.00007323709,0.0003291985,0.00005817474,0.0022833,0.000123877],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002279826,"about_ca_system_score_gemma":0.003376764,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00151273,"about_ca_topic_score_gemma":0.001428118,"domain_scores_codex":[0.9925616,0.004018455,0.0008468151,0.0007530103,0.0005448812,0.001275208],"domain_scores_gemma":[0.9979044,0.000541923,0.0002564757,0.000603336,0.0001947136,0.0004991377],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00002397547,0.0001899163,0.8449768,0.0005200887,0.00001898383,0.00001035347,0.00679447,0.00989558,4.025512e-8,0.0003061237,0.09285924,0.04440443],"study_design_scores_gemma":[0.0003006167,0.0008465761,0.04942351,0.001325986,0.000008677579,1.664946e-7,0.01305383,0.9005604,4.742896e-7,0.000572464,0.03365728,0.000249954],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1897873,0.006020512,0.3790211,0.3689652,0.034933,0.01522646,0.0003341792,0.002836772,0.002875507],"genre_scores_gemma":[0.9842606,0.0001638977,0.00865984,0.004791016,0.0005639782,0.0009270679,0.00007383949,0.00008278666,0.0004770074],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.8906649,"threshold_uncertainty_score":0.9999913,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.07253092882360608,"score_gpt":0.4257990989987154,"score_spread":0.3532681701751094,"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."}}