Effective hybrid feature selection using different bootstrap enhances cancers classification performance

Table 21 The proposed methods compared with the hybrid of MIFS and RFE

Datasets	Train Data %	Test Data %	Over-fitting Diff. %	Pre	Rec	F1-score	NO.F	F-Time (sec)	C-Time (sec)	AUC	Var.	ACC %
RF classifier
RNA gene	100.000	99.501	0.499	0.794	0.715	0.723	5000	10,227.579	1.199	1.000	0.000041	99.501
DNA CNV	92.908	85.034	7.874	0.770	0.716	0.717	4500	88,434.627	3.411	0.946	0.000698	85.034
Parkinson’s disease	100.000	83.861	16.139	0.809	0.737	0.759	150	74.445	0.411	0.876	0.002867	83.861
Dermatology diseases	99.727	94.819	4.908	0.941	0.930	0.930	12	1.113	0.079	0.996	0.001528	94.819
BreastEW	100.000	95.965	4.035	0.961	0.953	0.956	10	1.592	0.133	0.988	0.000756	95.965

ISSN: 1756-0381