Table 2.
Predictive performance comparison of the eight types of machine learning algorithms in the training and validation dataset.
| Methods | Training dataset | Validation dataset | ||||
|---|---|---|---|---|---|---|
| AUC | Sensitivity | Specificity | AUC | Sensitivity | Specificity | |
| LR | 0.744 | 0.615 | 0.771 | 0.693 | 0.881 | 0.433 |
| GBM | 0.878 | 0.692 | 0.937 | 0.858 | 0.742 | 0.837 |
| XGB | 0.907 | 0.762 | 0.934 | 0.849 | 0.682 | 0.865 |
| RF | 0.902 | 0.767 | 0.950 | 0.843 | 0.795 | 0.798 |
| DT | 0.692 | 0.659 | 0.724 | 0.652 | 0.603 | 0.680 |
| NNET | 0.889 | 0.692 | 0.945 | 0.811 | 0.656 | 0.837 |
| SVM | 0.771 | 0.541 | 0.876 | 0.750 | 0.642 | 0.764 |
| BN | 0.781 | 0.674 | 0.755 | 0.777 | 0.675 | 0.792 |
AUC, the area under the curve; LR, logistic regression; GBM, gradient boosting machine; XGB, extreme gradient boosting; RF, random forest; DT, decision tree; NNET, neural network; SVM, support vector machine; BN, Bayesian network.