Table II.
Classification effects of SVM method and RF method.
| Method | No. of samples | Correct rate | Sensitivity | Specificity | PPV | NPV | AUROC |
|---|---|---|---|---|---|---|---|
| SVM | |||||||
| Training group | 371 | 0.9757 | 0.9556 | 0.9822 | 0.9451 | 0.9857 | 0.997 |
| Test group | 488 | 0.9426 | 0.8913 | 0.9629 | 0.9044 | 0.9574 | 0.876 |
| Combined | 859 | 0.9430 | 0.8947 | 0.9604 | 0.8908 | 0.9619 | 0.902 |
| RF | |||||||
| Training group | 371 | 0.9650 | 0.9444 | 0.9715 | 0.9140 | 0.9820 | 0.983 |
| Test group | 488 | 0.9057 | 0.8478 | 0.9286 | 0.8239 | 0.9393 | 0.818 |
| Combined | 859 | 0.9267 | 0.9035 | 0.9350 | 0.8340 | 0.9656 | 0.916 |
PPV, positive predictive value; NPV, negative predictive value; AUROC, area under the receiver operating characteristic curve; SVM, support vector machine; RF, random forest.