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

Table 3

Performance analysis of classifiers in terms of PC 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	71.48688	87.5	79.69	52.27	52.27	59.89258
	60	70.44813	88.0225	86.72	51.625	78.38813	74.96797
	90	51.6875	63.8	79.17	52	57.42563	85.02938
Adaboost	30	68.75	71.3575	82.295	51.5	51.125	79.93242
	60	88.0225	87.11	91.15	55.00906	53.51875	71.875
	90	84.38	65.884	71.48688	52.4725	51.25	77.24688
LR	30	61.72	80.99	89.0675	54.55188	55.66219	86.97938
	60	53.35	73.50344	80.21	52.50625	51.21875	97.39375
	90	77.60625	79.17	53.73813	86.46	71.09875	88.06888
DT	30	94.79	54.10938	55.07438	91.41	84.38	88.47734
	60	91.15	78.38813	56.7725	55.12336	91.15	88.47734
	90	59.38	56.90313	89.59	87.5	95.31	65.73648
QDA	30	52.7425	54.16	87.11	72.65625	86.72	95.83
	60	77.085	83.86	88.545	54.10938	73.31953	91.67
	90	98.96	84.38	84.38	52.135	82.95589	91.40625
Average		73.43725	73.94254	78.33329	61.42191	69.05284	82.86558