From: A new regularized least squares support vector regression for gene selection
Procedures | Classifier | No. of genes | Accuracy |
---|---|---|---|
A: Proposed selection and criterion | |||
RLS-SVR | Â | Â | Â |
δ j ≥ 1/q | +KFDA | 5 | 0.9091 |
 | +SVM | 5 | 0.9545 |
∑ δ j ≥ 80% | +KFDA | 7 | 0.9091 |
 | +SVM | 7 | 0.9545 |
q genes | +KFDA | 10 | 0.9545 |
 | +SVM | 10 | 0.9545 |
B: Other selection procedures | |||
MBGS | +KFDA | 10 | 0.9545 |
 | +SVM | 10 | 1 |
SGS1 | +KFDA | 10 | 0.9091 |
 | +SVM | 10 | 0.9545 |
SGS2 | +KFDA | 10 | 0.9091 |
 | +SVM | 10 | 0.9545 |
C: Selection and classification together | |||
 | BMA | 13–18 | 0.7273 |
 | MBGS | 10 | 1 |