Identification of Tilletia foetida, Ustilago tritici, and Urocystis tritici Based on Near-Infrared Spectroscopy

Table 6

The optimal results except the average identification accuracies and the standard deviations of the identification accuracies for the 10 data subsets for the training sets and the testing sets were 100% and 0 when the DPLS models for Data treatment method 2 were built with different modeling ratios in each modeling spectral region.

Spectral region (cm^−1)	Modeling ratio of training set to testing set	The number of principal components	Training sets		Testing sets
			Average identification accuracy for the 10 data subsets (%)	Standard deviation of the identification accuracies for the 10 data subsets	Average identification accuracy for the 10 data subsets (%)	Standard deviation of the identification accuracies for the 10 data subsets
6000–7000	3 : 1	7	100	0	99.78	0.006660
8000–9000	3 : 1	6	100	0	99.78	0.006660
8000–9000	4 : 1	6	100	0	99.44	0.01112
8000–9000	5 : 1	6	100	0	99.67	0.009990
9000–10000	3 : 1	5	99.70	0.006994	98.45	0.01734
9000–10000	4 : 1	6	100	0	98.61	0.02241
9000–10000	5 : 1	5	99.66	0.01026	99.33	0.01332