Table 1.

The summary of classification measures, including accuracy, AUC (area under the curve), F1 scores, and precision scores, achieved by four different classification methods: linear discriminant analysis (LDA), support vector machine (SVM), K-nearest neighbour (KNN), and artificial neural network (ANN)

	Pre-treatment/Feature selection	CV-validation	Test	AUC		F1		Precision
LDA					A		B		A	B
	Raw	0.78	0.70	0.79	0.70		0.63		0.68	0.77
	SNV	0.83	0.90	0.91	0.79		0.72		0.76	0.86
	PCA(3pc)	0.70	0.90	0.79	0.67		0.54		0.63	0.79
ANN	Raw	0.80	0.70	0.85	0.76		0.78		0.80	1.00
	SNV	0.81	0.80	0.83	0.78		0.75		0.80	1.00
	PCA(3pc)	0.95	0.95	0.69	0.67		0.55		0.70	1.00
KNN	Raw	0.76	0.80	0.84	0.69		0.62		0.64	0.66
	SNV	0.78	0.70	0.81	0.74		0.68		0.76	1.00
	PCA(3pc)	0.73	0.90	0.80	0.72		0.59		0.70	0.95
SVM	Raw	0.83	0.60	0.89	0.86		0.83		0.92	1.00
	SNV	0.95	0.90	0.94	0.93		0.86		0.88	0.95
	PCA(3pc)	0.78	0.85	0.81	0.76		0.66		0.76	1.00

In this table, A and B refer to two different classes, respectively, for healthy sample and infested ones

Bold values demonstrate the best performance model