An Amalgamated Approach to Bilevel Feature Selection Techniques Utilizing Soft Computing Methods for Classifying Colon Cancer

Table 2

Performance analysis of classifiers in terms of classification accuracies with six optimization techniques for different gene selection methods using 30-60-90 selected genes.

Classifiers	Gene selection	Optimization techniques
		Invasive Weed Optimization	Teaching Learning-Based Optimization	League Championship Optimization	Beetle Antennae Search Optimization	Crow Search Optimization	Fruit fly Optimization
RF	30	85.74344	93.75	89.85625	76.135	76.135	79.94629
	60	85.22406	94.01125	93.36	76.3375	89.20625	87.48398
	90	75.84375	81.9	89.6	76	78.71281	92.51797
Adaboost	30	84.375	85.67875	91.1475	75.75	76.3375	89.96621
	60	94.01125	93.555	95.575	76.3375	76.75938	85.9375
	90	92.19	82.942	85.74344	76.23625	75.625	88.62695
LR	30	80.86	90.49688	94.53375	77.27594	77.83109	93.49766
	60	76.675	86.76016	90.1125	76.25313	75.60938	98.69688
	90	88.8125	89.6	76.86906	93.23	85.54938	94.03444
DT	30	97.395	77.05469	77.53719	95.705	92.19	94.23867
	60	95.575	89.20625	78.38625	76.3375	95.575	94.23867
	90	79.69	78.45156	94.795	93.75	97.655	82.86824
QDA	30	76.37125	77.08	93.555	86.32813	93.36	97.915
	60	88.55	91.93	94.2725	76.3375	86.66	95.835
	90	99.16	92.19	92.19	76.3375	91.47912	95.70313
Average		86.71975	86.97377	89.1689	80.55673	84.57899	91.43377