Skip to main content

Table 5 The AUROC values of the ensemble algorithms, resampling algorithms, and the ELAS

From: An ensemble learning with active sampling to predict the prognosis of postoperative non-small cell lung cancer patients

Task

Ensemble algorithms

Resampling algorithms

Proposed

SVM-AdaBoost

SVM-Bagging

SVM-SMOTE

SVM-TomekLinks

SVM-ELAS

Mean

SD

Mean

SD

Mean

SD

Mean

SD

Mean

SD

1-year recurrence

0.682

0.082

0.673

0.072

0.620

0.073

0.650

0.065

0.702

0.079

1-year death

0.768

0.055

0.726

0.047

0.670

0.058

0.668

0.058

0.760

0.042

3-year recurrence

0.692

0.038

0.723

0.037

0.706

0.031

0.723

0.038

0.728

0.033

3-year death

0.707

0.043

0.721

0.039

0.710

0.030

0.711

0.043

0.733

0.035

5-year recurrence

0.752

0.055

0.752

0.053

0.751

0.053

0.752

0.053

0.748

0.055

5-year death

0.724

0.031

0.739

0.032

0.732

0.031

0.738

0.036

0.742

0.029

All tasks

0.721

0.062

0.722

0.054

0.698

0.065

0.707

0.062

0.736

0.052

  1. The bold means the best results for corresponding tasks