Research Article
A Wrapper Feature Subset Selection Method Based on Randomized Search and Multilayer Structure
Table 4
Comparison of the average accuracy rates of MLFSSM with four feature selection methods.
| Method | Breast | Leukemia | Colon | Hepato | CNS |
| SVM-RFE | 0.877 | 0.967 | 0.835 | 0.658 | 0.693 | LS-bound | 0.778 | 0.935 | 0.817 | 0.618 | 0.61 | Bayes + KNN | 0.821 | 0.92 | 0.828 | 0.628 | 0.628 | EN-LR | 0.854 | 0.962 | 0.837 | 0.683 | 0.664 | GRRF | 0.846 | 0.939 | 0.834 | 0.67 | 0.618 | T-SS | 0.893 | 0.97 | 0.871 | 0.693 | 0.655 | MLFSSM | 0.959 | 0.969 | 0.961 | 0.943 | 0.843 |
|
|