Gene selection via improved nuclear reaction optimization algorithm for cancer classification in high-dimensional data

Journal of Big Data

Table 5 Fitness values of the proposed RBNRO-DE against popular ML classifiers

Benchmark	RBNRO-DE with k-NN	RBNRO-DE with SVM	k-NN	SVM	DT	RF	XGBoost
BLCA	0.0029	0.0025	0.0445	0.0215	0.0215	0.0676	0.0330
CESC	0.0185	0.0025	0.0260	0.0419	0.0260	0.0260	0.0260
CHOL	0.0025	0.0025	0.0100	0.0100	0.0100	0.0100	0.1200
COAD	0.0025	0.0025	0.0250	0.0250	0.0250	0.0250	0.0400
ESCA	0.0279	0.0113	0.0354	0.0354	0.0608	0.0354	0.0354
GBM	0.0025	0.0025	0.0409	0.0100	0.0100	0.0100	0.0100
HNSC	0.0089	0.0025	0.0538	0.1414	0.0626	0.0713	0.0626
KICH	0.0025	0.0025	0.0100	0.0100	0.0100	0.0100	0.0100
KIRC	0.0047	0.0025	0.0264	0.0345	0.1245	0.0836	0.0345
KIRP	0.0025	0.0025	0.0100	0.0100	0.0100	0.0100	0.0100
LIHC	0.0025	0.0026	0.0100	0.0100	0.0333	0.0449	0.0333
LUAD	0.0112	0.0103	0.0272	0.0272	0.0444	0.0186	0.0272
LUSC	0.0026	0.0025	0.0368	0.0278	0.1170	0.0368	0.0278
PAAD	0.0025	0.0025	0.0100	0.0100	0.0100	0.0100	0.0100
PCPG	0.0025	0.0025	0.0100	0.0100	0.0100	0.0100	0.0100
READ	0.0025	0.0025	0.0100	0.0100	0.0100	0.0100	0.0100
SARC	0.0212	0.0067	0.0287	0.0287	0.0287	0.0287	0.0287
SKCM	0.0025	0.0025	0.0100	0.0100	0.0571	0.0100	0.0100
STAD	0.0029	0.0125	0.0760	0.0760	0.1530	0.0980	0.0760
THCA	0.0025	0.0025	0.0100	0.0100	0.0454	0.0100	0.0100
THYM	0.0025	0.0025	0.0100	0.0100	0.0100	0.0100	0.0100
UCEC	0.0038	0.0025	0.0842	0.0842	0.0842	0.0595	0.0595
Ranking (\(\text {W\|T\|L}\))	2\|11\|9	\(\mathbf {9\|11\|2}\)	0\|0\|22	0\|0\|22	0\|0\|22	0\|0\|22	0\|0\|22