Table 4.

Classification accuracies on the validation and test sets for the multiclass classification among the 3 groups (healthy, mild, and severe symptoms) for the 3 data sets.

Data set		CATa	SVMb	LRc	P value	CAT vs SVM	CAT vs LR	SVM vs LR
Italian, mean (SD)
	Validation	0.67 (0.03)	0.64 (0.02)	0.63 (0.03)	.006	.46	.005	.12
	Test	0.54 (0.08)	0.57 (0.16)	0.59 (0.13)	N/Ad	N/A	N/A	N/A
Spanish, mean (SD)
	Validation	0.63 (0.02)	0.64 (0.02)	0.60 (0.02)	.002	.88	.02	.003
	Test	0.49 (0.09)	0.53 (0.11)	0.51 (0.15)	N/A	N/A	N/A	N/A
Italian&Spanish, mean (SD)
	Validation	0.61 (0.01)	0.58 (0.02)	0.58 (0.02)	.008	.01	.03	.92
	Test	0.53 (0.06)	0.54 (0.08)	0.52 (0.09)	N/A	N/A	N/A	N/A

^aCAT: CatBoost classifier.

^bSVM: support vector machine.

^cLR: logistic regression.

^dN/A: not applicable.