Table 4 Area under the ROC curve of parsimony models over seven benchmark data using four information theoretic filters
From: iRDA: a new filter towards predictive, stable, and enriched candidate genes
| Â | mRMR | CMIM | FCBF | iRDA | ||||
|---|---|---|---|---|---|---|---|---|
| Â | % | # | % | # | % | # | % | # |
Brain | 97.4 | 5 | 98.86 | 4 | 94.72 | 5 | 99.68 | 3 |
CNS | 87.18 | 4 | 89.68 | 4 | 91.15 | 5 | 94.08 | 5 |
Colon | 92.5 | 4 | 96.25 | 5 | 89.72 | 3 | 98.86 | 4 |
Leukemia | 100 | 4 | 100 | 4 | 99.96 | 3 | 100 | 3 |
Lung | 99.94 | 4 | 99.66 | 5 | 97.08 | 4 | 100 | 4 |
Lymphoma | 99 | 5 | 98.77 | 4 | 50.36 | 5 | 97.37 | 3 |
Prostate | 97.62 | 7 | 96.83 | 6 | 97.67 | 7 | 96.77 | 7 |
avg | 96.23 | 4.71 | 97.15 | 4.57 | 88.67 | 4.57 | 98.11 | 4.14 |