NMFBFS: A NMF-Based Feature Selection Method in Identifying Pivotal Clinical Symptoms of Hepatocellular Carcinoma

Table 7

Classification accuracy of inferred optimal feature subset via NMFBFS, ReliefF, mRMR, and Elastic Net on the training set.

Methods	Feature subset	Dimension	Classification accuracy in LSSVM (%)
NMFBFS		39	80.002 ± 9.95
ReliefF	FSRF20	20	65.00 ± 10.03
	FSRF40	40	73.33 ± 15.76
mRMR	FSMR20	20	70.83 ± 12.5
	FSMR40	40	74.17 ± 9.03
Elastic Net	FSEN20	20	70.00 ± 11.56
	FSEN40	40	76.67 ± 10.46