Table 2.
Performances of the seven machine learning models for predicting in-hospital mortality
| ML | Accuracy | AUC |
|---|---|---|
| XGBoost | ||
| Training set | 1.000 | 1.000 |
| Validation set | 0.895 | 0.884 |
| RF | ||
| Training set | 1.000 | 1.000 |
| Validation set | 0.891 | 0.882 |
| SVM | ||
| Training set | 0.875 | 0.747 |
| Validation set | 0.872 | 0.763 |
| LR | ||
| Training set | 0.882 | 0.833 |
| Validation set | 0.890 | 0.845 |
| NB | ||
| Training set | 0.856 | 0.836 |
| Validation set | 0.862 | 0.856 |
| KNN | ||
| Training set | 0.904 | 0.940 |
| Validation set | 0.868 | 0.651 |
| DT | ||
| Training set | 1.000 | 1.000 |
| Validation set | 0.842 | 0.655 |
ML machine learning, XGBoost eXtreme Gradient Boosting, RF Random Forest, SVM Support Vector Machine (radial bias function), LR Logistic Regression, NB Naive Bayes, KNN k-Nearest Neighbors, DT Decision Tree, AUC the area under curve