Protected Geographical Indication Identification of a Chinese Green Tea (Anji-White) by Near-Infrared Spectroscopy and Chemometric Class Modeling Techniques

Table 2

Predicting results obtained by OCPLS and SIMCA.

Model	Data treatment	TP^b	FN	TN	FP	Sensitivity	Specificity
OCPLS	Raw data (11^a)	34	10	68	13	0.773	0.840
	Smoothing (10)	36	8	65	16	0.818	0.803
	2nd derivative (9)	37	7	73	8	0.841	0.901
	SNV (8)	39	5	77	4	0.886	0.951
SIMCA	Raw data (10)	35	9	70	11	0.796	0.864
	Smoothing (8)	33	11	68	13	0.750	0.840
	2nd derivative (9)	36	8	74	7	0.818	0.914
	SNV (7)	39	5	76	5	0.886	0.938

^aNumbers of OCPLS or SIMCA latent variables.
^bTP: number of true positives; FN: number of false negatives; TN: number of true negatives; FP: number of false positives. “Positive” and “negative” represent Anji- and non-Anji-white tea, respectively.