The Feasibility of Using Near-Infrared Spectroscopy and Chemometrics for Untargeted Detection of Protein Adulteration in Yogurt: Removing Unwanted Variations in Pure Yogurt

Table 2

Predicting results of OCPLS models for pure and adulterated yogurt objects.

Preprocessing	LVs^a	Pure yogurt	A1	A2	A3	Sensitivity^b	Specificity^c
Raw spectra (𝜇=0.0998,𝜎=0.00644)^d	7	3^e	12	6	16	0.850 (17/20)	0.827 (163/197)
SNV (𝜇=1.0004,𝜎=0.00621)	5	2	3	5	8	0.900 (18/20)	0.919 (181/197)
OP (𝜇=1.0001,𝜎=0.00592)	5	2	2	3	5	0.900 (18/20)	0.949 (187/197)

 ^aThe number of PLS components.
^bThe numbers in the brackets denote TP/(TP + FN).
^cThe numbers in the brackets denote TN/(TN + FP).
^dModel parameters.
^eThe number of objects that were wrongly predicted.