Table 3 Medicare part D results comparison
From: Iterative cleaning and learning of big highly-imbalanced fraud data using unsupervised learning
Medicare part D | ||
|---|---|---|
Model | AUC | AUPRC |
AE-0 | 0.5181 | 0.0238 |
AE-1 | 0.5276 | 0.0350 |
AE-2 | 0.5547 | 0.0736 |
AE-3 | 0.5791 | 0.1185 |
AE-4 | 0.5906 | 0.1516 |
AE-5 | 0.5975 | 0.1778 |
AE-6 | 0.6016 | 0.2010 |
AE-7 | 0.6033 | 0.2198 |
AE-8 | 0.6116 | 0.2468 |
AE-9 | 0.6093 | 0.2598 |
IF | 0.5083 | 0.0176 |
COPOD | 0.5668 | 0.1179 |