mRMR-ABC: A Hybrid Gene Selection Algorithm for Cancer Classification Using Microarray Gene Expression Profiling

Table 6

Comparison between mRMR-ABC and ABC algorithms classification performance when applied with the SVM classifier for leukemia1 dataset.

Number of genes	Classification accuracy
	mRMR-ABC			ABC
	Best	Mean	Worst	Best	Mean	Worst
2	91.66%	89.63%	81.94%	87.5%	86.45%	81.94%
3	93.05%	90.37%	83.33%	88.88%	89.82%	83.33%
4	94.44%	91.29%	86.11%	88.8%	91.15%	83.33%
5	95.83%	92.82%	88.88%	91.66%	91.89%	87.5%
6	95.83%	92.82%	90.32%	91.99%	92.04%	87.5%
7	97.22%	93.10%	90.32%	93.05%	92.23%	87.5%
10	98.61%	94.44%	91.66%	93.05%	92.38%	88.88%
13	98.61%	94.93%	91.66%	93.05%	92.44%	88.88%
14	100%	95.83%	93.05%	93.05%	92.51%	88.88%