Fuzzy Weighted Least Squares Support Vector Regression with Data Reduction for Nonlinear System Modeling

Table 10

Comparison results of the proposed method, [15], G-LSSVR and Local-LSSVR with the hyper-parameter set , and the overlap factor as 2.8 are shown for Example 4.

: the number of the SRMs	RMSE (Training)		RMSE (Testing)
	our approach	[15]	our approach	[15]
4-SRMs	0.2544	0.1589	0.3304	0.5163
8-SRMS	0.2506	0.1365	0.3321	0.4861
12-SRMs	0.2491	0.1248	0.3338	0.4218
G-LSSVR	0.2480		0.3355
Local-LSSVR
M=41 training points	0.3328		0.3805
M=61 training points	0.3234		0.3666
M=81 training points	0.3237		0.3366