Table 3.
Average performance, standard deviation (shown in parenthesis) and p-value of different predications model based on 10-fold validation in 30 independent runs.
| Method | Accuracy | F1-Score | Sensitivity | Specificity | AUROC |
|---|---|---|---|---|---|
| XGBoost | 87.71 (1.34) | 71.45 (1.42) | 67.52 (1.38) | 90.82 (1.26) | 89.36 (1.24) |
| SVM | 84.79 (1.31) | 72.48(1.71) | 67.01 (1.36) | 88.96 (0.67) | 87.69 (1.31) |
| MLP | 85.25 (1.29) | 69.92 (1.37) | 62.21 (1.03) | 88.17 (1.43) | 88.38 (1.36) |
| XDT | 88.41 (0.59) | 76.17 (0.67) | 67.21 (0.82) | 91.02 (0.69) | 90.62 (1.03) |
| Proposed Model | 89.97 (1.08) | 78.13 (1.21) | 71.65 (0.76) | 92.99 (0.82) | 92.81 (1.06) |
| P-value | 0.0034218 | 0.0037548 | 0.003295 | 0.004606 | 0.004438 |