From: Data reduction techniques for highly imbalanced medicare Big Data
Features classifier | 10 | 15 | 20 | 25 | 30 | 82 |
---|---|---|---|---|---|---|
CatBoost | 0.7589 | 0.8051 | 0.7973 | 0.7984 | 0.7938 | 0.7798 |
ET | 0.4986 | 0.4151 | 0.3933 | 0.4009 | 0.3668 | 0.2401 |
LightGBM | 0.7070 | 0.7400 | 0.7100 | 0.7019 | 0.6937 | 0.6783 |
Logistic regression | 0.3117 | 0.3189 | 0.3117 | 0.3170 | 0.2486 | 0.2700 |
Random forest | 0.4820 | 0.4753 | 0.3983 | 0.4120 | 0.3545 | 0.2199 |
XGBoost | 0.7473 | 0.7860 | 0.7588 | 0.7554 | 0.7491 | 0.7351 |