Explainable machine learning models for Medicare fraud detection

Journal of Big Data

Table 11 Mean AUPRC values by classifier and number of features for ten iterations of fivefold cross validation, for classifying the Medicare Part B data

Features classifier	10	15	20	25	30	80
CatBoost	0.6581	0.6792	0.7069	0.7009	0.7016	0.6817
ET	0.0400	0.0462	0.0443	0.0524	0.0424	0.0433
LightGBM	0.3939	0.3830	0.4261	0.4589	0.4293	0.4146
Logistic Regression	0.0093	0.0326	0.0338	0.0065	0.0064	0.0103
Random Forest	0.4356	0.3990	0.3736	0.3800	0.3395	0.2462
XGBoost	0.6611	0.6715	0.6995	0.6956	0.6955	0.6886