Feature Selection Method Based on Partial Least Squares and Analysis of Traditional Chinese Medicine Data

Table 7

Comparison of experimental results of the LAPLS with other methods (evaluation indicators: R² and RMSE).

	PLS		Lasso		PLSRFE		LAPLS
	R^2	RMSE	R^2	RMSE	R^2	RMSE	R^2	RMSE
WYHXB	0.5660	422.1680	0.4538	427.7071	0.6498	418.3305	0.6558	412.7325
NYWZ	0.6072	154.8713	0.6254	152.8729	0.6791	158.7410	0.7326	140.5172
DCQT	0.8262	0.01620	0.7988	0.0129	0.8848	0.0181	0.9384	0.0117
CCrime	0.6419	0.1388	0.6609	0.1306	0.7355	0.1413	0.6703	0.1516
BreastData	0.6333	3.5338	0.6414	3.1766	0.5777	3.6686	0.7064	3.1468
RBuild	0.9616	221.2931	0.9815	226.7571	0.9746	190.4369	0.9831	202.5260
Average	0.7060	133.6702	0.6936	135.1095	0.7502	128.5561	0.7811	126.5143