{"id":"W4412017252","doi":"10.18280/isi.300505","title":"SMOTE-ENN Resampling to Optimize Diabetes Prediction in Imbalanced Data","year":2025,"lang":"fr","type":"article","venue":"Ingénierie des systèmes d information","topic":"Artificial Intelligence in Healthcare","field":"Health Professions","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Resampling; Computer science; Artificial intelligence; Machine learning; Data mining; Pattern recognition (psychology)","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":true,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow","insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.003797101,0.0003857605,0.0006059785,0.001011641,0.001059833,0.0002107807,0.0009485703,0.0009050103,0.0003575587],"category_scores_gemma":[0.007752711,0.0004470147,0.00006746114,0.002081446,0.0002401146,0.005199983,0.0008834469,0.001323619,0.001364108],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002696729,"about_ca_system_score_gemma":0.001135476,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.006348702,"about_ca_topic_score_gemma":0.002947831,"domain_scores_codex":[0.9939529,0.0008909898,0.002900765,0.0005044909,0.0004241588,0.001326744],"domain_scores_gemma":[0.9955449,0.001333321,0.0006053525,0.001433387,0.0007998095,0.0002832956],"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":[0.0004640557,0.0001587047,0.2849424,0.0128082,0.0001201451,0.000005899586,0.07519262,0.04381509,0.0001532611,0.01408642,0.04993751,0.5183157],"study_design_scores_gemma":[0.0008624317,0.0002890043,0.0679967,0.02383864,0.00009589507,0.000002921893,0.04495483,0.7342841,0.0006960164,0.01758624,0.10859,0.0008031692],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.7750922,0.005019901,0.1369653,0.02263851,0.02314607,0.007842693,0.002945085,0.0007158228,0.02563437],"genre_scores_gemma":[0.9758592,0.0005550413,0.01377877,0.004402443,0.0007110573,0.0005090471,0.001710796,0.00004572025,0.002427906],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.690469,"threshold_uncertainty_score":0.9997982,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0928380235529106,"score_gpt":0.399481683025284,"score_spread":0.3066436594723734,"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."}}