Improving the Classification Accuracy for Near-Infrared Spectroscopy of Chinese Salvia miltiorrhiza Using Local Variable Selection

Table 3

Comparison of classification rates between different methods.

Set	Correct identified rate
	Full spectrum		Traditional VS		L-VS
	SIMCA	PLS-DA	SIMCA	PLS-DA	SIMCA	PLS-DA
Class 1	0.93	0.79	0.93	0.93	1.00	1.00
Class 2	0.80	0.60	0.80	0.80	1.00	1.00
Class 3	1.00	1.00	1.00	1.00	1.00	1.00
Class 4	0.40	1.00	0.40	1.00	1.00	1.00
Class 5	0.40	1.00	0.40	1.00	1.00	1.00
Class 6	0.71	0.86	0.57	0.71	1.00	1.00
Class 7	0.25	0.50	0.25	0.50	1.00	1.00
Class 8	0.57	0.71	0.57	0.86	1.00	1.00
Class 9	0.90	0.90	0.90	0.90	1.00	1.00
Class 10	0.75	1.00	0.75	1.00	1.00	1.00
Class 11	0.50	0.50	0.50	1.00	1.00	1.00
Class 12	0.75	1.00	0.75	1.00	1.00	1.00
Class 13	0.50	1.00	0.50	1.00	0.75	1.00

VS: variable selection; L-VS: local variable selection; SIMCA: soft independent modelling of class analogy; PLS-DA: partial least squares discriminant analysis.