Feature Selection of the Rich Model Based on the Correlation of Feature Components

Table 1

The performances of the original Dix2_T2 submodel of CC-JRM and the subfeatures selected by different methods.

Payload	(260-D)	Random		Proposed
		Number of feature components		Number of feature components
0.1	0.5033	88	0.5020	88	0.5046
0.2	0.5073	87	0.5065	87	0.5082
0.3	0.5167	72	0.5098	72	0.5151
0.4	0.5287	70	0.5189	70	0.5279
0.5	0.5469	86	0.5324	86	0.5491