Table 9 Comparison of SHAP and CatBoost feature selection methods in terms of their AUPRC scores
From: Feature selection strategies: a comparative analysis of SHAP-value and importance-based methods
Size | SHAP-CatBoost | CatBoost | p-value | Winner |
|---|---|---|---|---|
3 | 0.6106 | 0.7235 | 0.0000 | CatBoost |
5 | 0.7266 | 0.7745 | 0.0000 | CatBoost |
7 | 0.7897 | 0.8279 | 0.0000 | CatBoost |
10 | 0.8333 | 0.8472 | 0.0000 | CatBoost |
15 | 0.8506 | 0.8491 | 0.7502 | Tie |