Table 1.
A comparison of the proposed predictor with the individual composition-based SVM model on training dataset.
| Methods | MCC | Accuracy | Sensitivity | Specificity | AUC | P-value |
|---|---|---|---|---|---|---|
| PVP-SVM | 0.695 | 0.870 | 0.737 | 0.933 | 0.900 | |
| SVM control | 0.554 | 0.811 | 0.636 | 0.894 | 0.837 | 0.068 |
| AAC | 0.525 | 0.792 | 0.841 | 0.687 | 0.841 | 0.086 |
| DPC | 0.395 | 0.743 | 0.837 | 0.546 | 0.760 | 0.00023 |
| CTD | 0.534 | 0.801 | 0.880 | 0.636 | 0.819 | 0.022 |
| DPC | 0.478 | 0.782 | 0.889 | 0.556 | 0.812 | 0.014 |
| ATC | 0.252 | 0.708 | 0.091 | 1.000 | 0.788 | 0.002 |
The first column represents the method name employed in this study. The second, the third, the fourth and the fifth respectively represent the MCC, accuracy, sensitivity, and specificity. The sixth column and the seventh represent the AUC and pairwise comparison of ROC area under curves (AUCs) between PVP-SVM and the other methods using a two-tailed t-test. A P ≤ 0.05 indicates a statistically meaningful difference between PVP-SVM and the selected method (shown in bold italic).