Table 2.
Comparison of results among different machine learning algorithms.
| DataSet | Algorithms | AUC (%) | Specificity(%) | Sensitivity(%) |
|---|---|---|---|---|
| Training set | Logistic regression | 96.94 | 93.76 | 80.35 |
| Random Forest | 99.99 | 99.99 | 99.99 | |
| GBDT | 99.46 | 98.83 | 90.32 | |
| Adaboost | 98.35 | 96.18 | 88.86 | |
| XGBoost | 99.99 | 99.99 | 99.99 | |
| CatBoost | 99.81 | 98.12 | 94.51 | |
| Test set | Logistic regression | 95.14 | 90.92 | 77.27 |
| Random Forest | 93.24 | 91.27 | 78.58 | |
| GBDT | 93.17 | 90.23 | 72.73 | |
| Adaboost | 93.15 | 91.15 | 72.73 | |
| XGBoost | 94.28 | 91.19 | 77.27 | |
| CatBoost | 95.34 | 93.17 | 77.27 |
The bold values means the highest value.