Explainable machine learning models for Medicare fraud detection

Journal of Big Data

Table 9 Mean AUC values by classifier and number of features for ten iterations of fivefold cross validation, for classifying the Medicare Part D data (Part 1)

Features classifier	7a	7b	8	9	10
CatBoost	0.9223	0.9231	0.9228	0.9293	0.9383
ET	0.8541	0.8415	0.8448	0.8614	0.8574
LightGBM	0.7541	0.7449	0.7730	0.7904	0.8298
Logistic Regression	0.8850	0.8698	0.8816	0.8832	0.8869
Random Forest	0.8340	0.8222	0.8384	0.8244	0.8332
XGBoost	0.9276	0.9282	0.9278	0.9346	0.9408