Feature Selection Based on BP Neural Network and Adaptive Particle Swarm Algorithm

Table 1

Average accuracy of different data sets.

Item	PalmData25	LIHC	PRAD	t-test	Wilcoxon test	Original feature
BP-PSO	1.09	1.91	1.55	1.78	1.05	1.99
DTree	1.72	1.09	2.33	1.55	3.57	2.76
NDFS	4.43	4.22	5.01	4.37	5.33	5.9
RF	1.21	3.41	2.59	4.61	2.14	5.44
BP-PSO	1.42	2.64	2.54	3.42	4.75	1.81
THCA	6.24	3.35	5.85	1.12	5.51	3.33