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

Table 7

Comparison results of the proposed method, [15], G-LSSVR and Local-LSSVR with the hyper-parameter set , and the overlap factor as 1.5 are shown for Example 3.

: the number of the SRMs	RMSE (Training)		RMSE (Testing)
	our approach	[15]	our approach	[15]
3-SRMs	0.2523	0.2715	0.2546	0.2832
5-SRMS	0.2531	0.2659	0.2661	0.2762
7-SRMs	0.2287	0.2487	0.2537	0.2645
9-SRMs	0.2298	0.2465	0.2464	0.2583
G-LSSVR	0.2922		0.3185
Local-LSSVR
M=21 training points	1.8583		1.8721
M=41 training points	0.2332		0.3383
M=61 training points	0.2259		0.3267
M=81 training points	0.2433		0.3561