{"id":"W4412460893","doi":"10.2196/75117","title":"A Machine Learning Algorithm With an Oversampling Technique in Limited Data Scenarios for the Prediction of Present and Future Restorative Treatment Need: Development and Validation Study","year":2025,"lang":"en","type":"article","venue":"JMIR Medical Informatics","topic":"Dental Health and Care Utilization","field":"Dentistry","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Preprint; Computer science; Oversampling; Machine learning; Artificial intelligence; Data mining; Algorithm; World Wide Web","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.01731984,0.001309308,0.00132322,0.001603154,0.0006044895,0.0008989263,0.001526708,0.001593431,0.001027035],"category_scores_gemma":[0.02092955,0.0004406103,0.001733893,0.0008999346,0.0005158964,0.001069971,0.001168666,0.001694092,0.0003819007],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0009117976,"about_ca_system_score_gemma":0.001711224,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.006661919,"about_ca_topic_score_gemma":0.003429289,"domain_scores_codex":[0.995871,0.00257161,0.0003310935,0.0005221101,0.0004709316,0.0002332134],"domain_scores_gemma":[0.9831092,0.01206367,0.0006031906,0.001146442,0.002810657,0.0002667603],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.001391896,0.002257542,0.07616483,0.0003714457,0.0007998093,0.0002248051,0.0003579818,0.6061345,0.002719003,0.001427338,0.002121633,0.3060293],"study_design_scores_gemma":[0.00004927588,0.0004867779,0.004589766,0.00003356518,0.00005054362,0.0000388701,0.00004679106,0.9934335,0.0007062952,0.0002476661,0.0003060136,0.00001101549],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.7479563,0.001733575,0.2445464,0.0004281814,0.0002609461,0.001311277,0.0004593151,0.00108831,0.002215667],"genre_scores_gemma":[0.8449938,0.000358299,0.1523641,0.0001264302,0.00004764907,0.0006391486,0.0007763846,0.00003271283,0.0006614606],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.01731984,"threshold_uncertainty_score":0.0915972,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03757868073169097,"score_gpt":0.3525947591697047,"score_spread":0.3150160784380137,"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."}}