Table 1.
Performance comparison for binary-class datasets.
| Classifier+Selection criterion | Number of genes | ||||
| 10 | 20 | 30 | 50 | 100 | |
| Colon cancer | |||||
| NMC+S2N | 12.2 ± 0.6 | 12.5 ± 0.6 | 12.3 ± 0.6 | 12.8 ± 0.6 | 12.9 ± 0.5 |
| NMC+MMC-RFE(U) | 13.9 ± 0.6 | 12.5 ± 0.6 | 12.5 ± 0.6 | 12.1 ± 0.6 | 11.2 ± 0.5 |
| NMC+MMC-RFE(O) | 13.4 ± 0.6 | 11.7 ± 0.6 | 11.5 ± 0.6 | 11.3 ± 0.6 | 11.2 ± 0.6 |
| NMC+SVM-RFE(H) | 16.2 ± 0.7 | 14.6 ± 0.6 | 13.7 ± 0.6 | 12.5 ± 0.6 | 11.6 ± 0.6 |
| NMC+SVM-RFE(S) | 13.3 ± 0.6 | 11.2 ± 0.6 | 10.7 ± 0.5 | 10.5 ± 0.5 | 10.9 ± 0.5 |
| MMC+MMC-RFE(U) | 13.6 ± 0.6 | 12.1 ± 0.6 | 11.9 ± 0.6 | 11.7 ± 0.5 | 11.0 ± 0.5 |
| MMC+MMC-RFE(O) | 13.2 ± 0.6 | 11.7 ± 0.6 | 11.5 ± 0.6 | 11.0 ± 0.6 | 11.1 ± 0.6 |
| SVM+SVM-RFE(H) | 18.3 ± 0.6 | 16.2 ± 0.7 | 15.8 ± 0.7 | 15.0 ± 0.6 | 15.0 ± 0.6 |
| SVM+SVM-RFE(S) | 13.5 ± 0.5 | 10.7 ± 0.5 | 10.2 ± 0.5 | 10.0 ± 0.5 | 10.5 ± 0.6 |
| Prostate cancer | |||||
| NMC+S2N | 10.1 ± 0.4 | 11.3 ± 0.5 | 12.3 ± 0.6 | 13.6 ± 0.6 | 16.0 ± 0.7 |
| NMC+MMC-RFE(U) | 9.9 ± 0.5 | 10.4 ± 0.5 | 10.8 ± 0.5 | 11.6 ± 0.5 | 13.4 ± 0.6 |
| NMC+MMC-RFE(O) | 9.6 ± 0.5 | 9.9 ± 0.5 | 10.3 ± 0.6 | 11.2 ± 0.6 | 13.5 ± 0.7 |
| NMC+SVM-RFE(H) | 9.6 ± 0.4 | 10.1 ± 0.5 | 10.2 ± 0.5 | 10.8 ± 0.5 | 12.2 ± 0.6 |
| NMC+SVM-RFE(S) | 9.7 ± 0.4 | 9.6 ± 0.4 | 10.0 ± 0.5 | 10.7 ± 0.5 | 12.4 ± 0.6 |
| MMC+MMC-RFE(U) | 8.8 ± 0.4 | 8.4 ± 0.4 | 8.4 ± 0.4 | 8.4 ± 0.4 | 8.6 ± 0.4 |
| MMC+MMC-RFE(O) | 8.5 ± 0.4 | 8.2 ± 0.4 | 7.9 ± 0.4 | 7.9 ± 0.4 | 8.1 ± 0.5 |
| SVM+SVM-RFE(H) | 9.9 ± 0.5 | 9.1 ± 0.4 | 9.3 ± 0.4 | 9.2 ± 0.4 | 9.1 ± 0.4 |
| SVM+SVM-RFE(S) | 8.5 ± 0.4 | 8.0 ± 0.4 | 8.5 ± 0.4 | 8.4 ± 0.4 | 8.8 ± 0.4 |
| Leukemia | |||||
| NMC+S2N | 5.6 ± 0.7 | 5.8 ± 0.6 | 5.4 ± 0.6 | 3.8 ± 0.5 | 3.2 ± 0.5 |
| NMC+MMC-RFE(U) | 5.7 ± 0.6 | 3.9 ± 0.5 | 3.8 ± 0.5 | 2.2 ± 0.4 | 0.8 ± 0.2 |
| NMC+MMC-RFE(O) | 5.8 ± 0.6 | 2.8 ± 0.5 | 1.8 ± 0.4 | 0.8 ± 0.2 | 0.4 ± 0.2 |
| NMC+SVM-RFE(H) | 5.4 ± 0.6 | 3.8 ± 0.5 | 3.4 ± 0.5 | 1.8 ± 0.4 | 0.6 ± 0.2 |
| NMC+SVM-RFE(S) | 6.0 ± 0.6 | 3.1 ± 0.4 | 2.0 ± 0.4 | 1.5 ± 0.3 | 0.9 ± 0.3 |
| MMC+MMC-RFE(U) | 5.6 ± 0.6 | 3.7 ± 0.5 | 3.7 ± 0.5 | 2.3 ± 0.4 | 0.8 ± 0.3 |
| MMC+MMC-RFE(O) | 5.8 ± 0.6 | 2.8 ± 0.5 | 1.6 ± 0.3 | 0.6 ± 0.2 | 0.3 ± 0.2 |
| SVM+SVM-RFE(H) | 4.1 ± 0.5 | 3.0 ± 0.4 | 2.9 ± 0.4 | 1.3 ± 0.3 | 1.3 ± 0.3 |
| SVM+SVM-RFE(S) | 3.8 ± 0.5 | 3.1 ± 0.4 | 2.5 ± 0.4 | 1.3 ± 0.3 | 1.3 ± 0.3 |
The average error and standard error rates (%) for Colon cancer, Prostate cancer and Leukemia, when the number of genes is {10, 20, 30, 50, 100}. SVM-RFE(S) shows the best result with respect to the C parameter; NMC+SVM-RFE(S): C = 0.01, SVM+SVM-RFE(S): C = 0.01 for Colon cancer; NMC+SVM-RFE(S): C = 0.01, SVM+SVM-RFE(S): C = 0.01 for Prostate cancer; NMC+SVM-RFE(S): C = 0.001, SVM+SVM-RFE(S): C = 100 for Leukemia.