{"id":"W4415127930","doi":"10.3390/risks13100198","title":"Application of Standard Machine Learning Models for Medicare Fraud Detection with Imbalanced Data","year":2025,"lang":"en","type":"article","venue":"Risks","topic":"Imbalanced Data Classification Techniques","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"Columbia College","funders":"","keywords":"Interpretability; Random forest; AdaBoost; Decision tree; Feature selection; Resampling; Preprocessor; Feature (linguistics); Ensemble learning","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":[0.01000176,0.001419966,0.00112541,0.002738194,0.000829966,0.002006695,0.001648999,0.001261251,0.0008552354],"category_scores_gemma":[0.02231446,0.0004570041,0.001260019,0.002431381,0.0005711211,0.002057519,0.001402589,0.002057756,0.0005617255],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002128761,"about_ca_system_score_gemma":0.002127307,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01090869,"about_ca_topic_score_gemma":0.01001837,"domain_scores_codex":[0.9964959,0.001744155,0.0002967834,0.0004898141,0.0007535073,0.0002199317],"domain_scores_gemma":[0.9914126,0.004568958,0.0007432917,0.001175163,0.001971936,0.0001280085],"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":[0.0003714558,0.0005340125,0.02568859,0.0001823503,0.0003308974,0.0001673799,0.0002257958,0.7724036,0.001341927,0.004908962,0.0036102,0.1902349],"study_design_scores_gemma":[0.000007031934,0.00004089811,0.001127098,0.00001544832,0.000009599231,0.00002231133,0.00002991913,0.9952551,0.0006030524,0.002421943,0.0004588147,0.000008850398],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.3461356,0.001644642,0.6412605,0.0019641,0.0002691752,0.0007265777,0.001478142,0.002362936,0.004158218],"genre_scores_gemma":[0.8233,0.0004778216,0.1730407,0.0003087885,0.00008338917,0.0003872733,0.001211983,0.00006435942,0.001125597],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.01090869,"threshold_uncertainty_score":0.05289501,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04827192005258184,"score_gpt":0.3258992892147127,"score_spread":0.2776273691621309,"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."}}