Table 10.
Comparison of best prediction accuracy for the colon tumor dataset.
| Methods (feature selection + classification) | # Selected genes | # Correctly-classified samples (accuracy) |
|---|---|---|
| α depended degree + decision rules | 1 | 52 (84%) |
| [this work] | 2 | 56 (90%) |
| HykGene + k-NNs, SVMs, C4.5, NB107 | 3 | 57 (92%) |
| MAVE + logistic discrimination108 | 50 | 52 (84%) |
| Clustering and rough sets attribute reduction + k-NNs109 | 6 | 49 (79%) |
| Clustering and rough sets attribute reduction + NB109 | 6 | 51 (82%) |
| Clustering and rough sets attribute reduction + C5.0109 | 6 | 56 (90%) |
| MRMR + NB110 | 9 | 58 (94%) |
| RBF + C4.5111 | 4 | 58 (94%) |
| ReliefF + C4.5111 | 4 | 53 (85%) |
| CFS-SF + C4.5111 | 26 | 55 (89%) |