Table 3 The results of performance evaluation with test sets of classification models for determining the geographical origin of domestic and imported kimchi samples.

From: Rapid determination of the geographical origin of kimchi by Fourier transform near-infrared spectroscopy coupled with chemometric techniques

Model¹⁾	Preprocessing method²⁾	Accuracy	Recall	Specificity	Precision	F1 score
KNN	Raw	1.00	1.00	1.00	1.00	1.00
	MSC	1.00	1.00	1.00	1.00	1.00
	MSC + D1	1.00	1.00	1.00	1.00	1.00
	MSC + D1 + ND	1.00	1.00	1.00	1.00	1.00
	MSC + D1 + SG	1.00	1.00	1.00	1.00	1.00
	MSC + D2	1.00	1.00	1.00	1.00	1.00
	MSC + D2 + ND	1.00	1.00	1.00	1.00	1.00
	MSC + D2 + SG	1.00	1.00	1.00	1.00	1.00
	SNV	1.00	1.00	1.00	1.00	1.00
	SNV + D1	1.00	1.00	1.00	1.00	1.00
	SNV + D1 + ND	1.00	1.00	1.00	1.00	1.00
	SNV + D1 + SG	1.00	1.00	1.00	1.00	1.00
	SNV + D2	1.00	1.00	1.00	1.00	1.00
	SNV + D2 + ND	1.00	1.00	1.00	1.00	1.00
	SNV + D2 + SG	1.00	1.00	1.00	1.00	1.00
CART	Raw	0.82	0.93	0.70	0.76	0.84
	MSC	0.97	1.00	0.93	0.94	0.97
	MSC + D1	0.98	1.00	0.97	0.97	0.98
	MSC + D1 + ND	0.98	1.00	0.97	0.97	0.98
	MSC + D1 + SG	0.93	0.97	0.90	0.91	0.94
	MSC + D2	0.82	0.73	0.90	0.88	0.80
	MSC + D2 + ND	0.98	1.00	0.97	0.97	0.98
	MSC + D2 + SG	0.90	0.87	0.93	0.93	0.90
	SNV	0.97	0.97	0.97	0.97	0.97
	SNV + D1	0.98	1.00	0.97	0.97	0.98
	SNV + D1 + ND	0.98	1.00	0.97	0.97	0.98
	SNV + D1 + SG	0.97	1.00	0.93	0.94	0.97
	SNV + D2	0.75	0.73	0.77	0.76	0.75
	SNV + D2 + ND	0.97	0.97	0.97	0.97	0.97
	SNV + D2 + SG	0.88	0.80	0.97	0.96	0.87
SVM	Raw	1.00	1.00	1.00	1.00	1.00
	MSC	1.00	1.00	1.00	1.00	1.00
	MSC + D1	1.00	1.00	1.00	1.00	1.00
	MSC + D1 + ND	1.00	1.00	1.00	1.00	1.00
	MSC + D1 + SG	1.00	1.00	1.00	1.00	1.00
	MSC + D2	0.98	1.00	0.97	0.97	0.98
	MSC + D2 + ND	1.00	1.00	1.00	1.00	1.00
	MSC + D2 + SG	1.00	1.00	1.00	1.00	1.00
	SNV	1.00	1.00	1.00	1.00	1.00
	SNV + D1	1.00	1.00	1.00	1.00	1.00
	SNV + D1 + ND	1.00	1.00	1.00	1.00	1.00
	SNV + D1 + SG	1.00	1.00	1.00	1.00	1.00
	SNV + D2	0.98	1.00	0.97	0.97	0.98
	SNV + D2 + ND	1.00	1.00	1.00	1.00	1.00
	SNV + D2 + SG	1.00	1.00	1.00	1.00	1.00
NB	Raw	0.65	0.77	0.53	0.62	0.69
	MSC	0.77	0.70	0.83	0.81	0.75
	MSC + D1	0.90	0.90	0.90	0.90	0.90
	MSC + D1 + ND	0.92	0.93	0.90	0.90	0.92
	MSC + D1 + SG	0.90	0.90	0.90	0.90	0.90
	MSC + D2	0.92	0.90	0.93	0.93	0.92
	MSC + D2 + ND	0.72	0.43	1.00	1.00	0.60
	MSC + D2 + SG	0.50	0.00	1.00	ID³⁾	ID
	SNV	0.73	0.67	0.80	0.77	0.71
	SNV + D1	0.90	0.90	0.90	0.90	0.90
	SNV + D1 + ND	0.92	0.93	0.90	0.90	0.92
	SNV + D1 + SG	0.90	0.90	0.90	0.90	0.90
	SNV + D2	0.93	0.93	0.93	0.93	0.93
	SNV + D2 + ND	0.93	0.90	0.97	0.96	0.93
	SNV + D2 + SG	0.93	0.90	0.97	0.96	0.93
RF	Raw	0.97	0.97	0.97	0.97	0.97
	MSC	0.98	0.97	1.00	1.00	0.98
	MSC + D1	0.97	0.97	0.97	0.97	0.97
	MSC + D1 + ND	1.00	1.00	1.00	1.00	1.00
	MSC + D1 + SG	0.98	1.00	0.97	0.97	0.98
	MSC + D2	0.92	0.90	0.93	0.93	0.92
	MSC + D2 + ND	0.98	1.00	0.97	0.97	0.98
	MSC + D2 + SG	0.88	0.90	0.87	0.87	0.89
	SNV	1.00	1.00	1.00	1.00	1.00
	SNV + D1	0.97	0.97	0.97	0.97	0.97
	SNV + D1 + ND	1.00	1.00	1.00	1.00	1.00
	SNV + D1 + SG	0.97	1.00	0.93	0.94	0.97
	SNV + D2	0.93	0.93	0.93	0.93	0.93
	SNV + D2 + ND	0.97	0.97	0.97	0.97	0.97
	SNV + D2 + SG	0.95	0.93	0.97	0.97	0.95
PLS-DA	Raw	0.90	0.87	0.93	0.93	0.90
	MSC	1.00	1.00	1.00	1.00	1.00
	MSC + D1	1.00	1.00	1.00	1.00	1.00
	MSC + D1 + ND	1.00	1.00	1.00	1.00	1.00
	MSC + D1 + SG	1.00	1.00	1.00	1.00	1.00
	MSC + D2	0.97	0.97	0.97	0.97	0.97
	MSC + D2 + ND	1.00	1.00	1.00	1.00	1.00
	MSC + D2 + SG	1.00	1.00	1.00	1.00	1.00
	SNV	1.00	1.00	1.00	1.00	1.00
	SNV + D1	1.00	1.00	1.00	1.00	1.00
	SNV + D1 + ND	1.00	1.00	1.00	1.00	1.00
	SNV + D1 + SG	1.00	1.00	1.00	1.00	1.00
	SNV + D2	0.97	0.97	0.97	0.97	0.97
	SNV + D2 + ND	1.00	1.00	1.00	1.00	1.00
	SNV + D2 + SG	1.00	1.00	1.00	1.00	1.00

¹⁾ KNN, k-nearest neighbors; CART, classification and regression tree; SVM, support vector machine; NB, naïve Bayes; RF, random forest; PLS-DA, partial least squares-discriminant analysis.
²⁾ MSC, multiplicative signal correction; D1, the first derivative; ND, Norris derivative filtering; SG, Savitzky-Golay filtering; D2, the second derivative; SNV, standard normal variate.
³⁾ ID, Indeterminate.

Back to article page

Table 3 The results of performance evaluation with test sets of classification models for determining the geographical origin of domestic and imported kimchi samples.

Search

Quick links