Research Article
mRMR-ABC: A Hybrid Gene Selection Algorithm for Cancer Classification Using Microarray Gene Expression Profiling
Table 5
Comparison between mRMR-ABC and ABC algorithms classification performance when applied with the SVM classifier for colon dataset.
| Number of genes | Classification accuracy | mRMR-ABC | ABC | Best | Mean | Worst | Best | Mean | Worst |
| 3 | 88.71% | 87.50% | 85.48% | 87.10% | 85.91% | 83.87% | 4 | 90.23% | 88.27% | 87.10% | 87.10% | 86.71% | 85.48% | 5 | 91.94% | 89.50% | 87.10% | 90.32% | 87.98% | 85.48% | 6 | 91.94% | 90.12% | 87.10% | 90.32% | 88.44% | 85.48% | 7 | 993.55% | 91.64% | 88.81% | 91.94% | 90.20% | 88.81% | 8 | 93.55% | 91.80% | 88.81% | 91.94% | 90.61% | 88.81% | 9 | 93.55% | 92.11% | 90.16% | 91.94% | 90.95% | 88.81% | 10 | 93.55% | 92.74% | 90.16% | 93.55% | 91.31% | 88.81% | 15 | 96.77% | 93.60% | 91.93% | 93.55% | 91.38% | 90.32% | 20 | 96.77% | 94.17% | 91.93% | 95.61% | 92.44% | 90.32% |
|
|