TABLE 2.

Results of multi-classification models of PKU. And, the bold values represent better results than other models.

	Models	Accuracy (%)	Sensitivity (%)	Specificity (%)	PPV (%)	AUC (%)
Training	RF	99.39	100.00	99.38	57.39	99.94
	MLP	99.88	100.00	99.88	87.42	99.96
	DT	99.59	100.00	99.59	67.00	99.97
	SGD	98.28	96.97	98.29	32.08	99.24
	LR	99.13	100.00	99.12	48.71	99.85
	KNN	100.00	100.00	100.00	100.00	100.00
Validation	RF	99.29	100.00	99.28	58.48	99.92
	MLP	99.71	89.86	99.81	82.67	99.87
	DT	99.46	98.55	99.47	65.39	99.14
	SGD	97.83	100.00	97.81	31.65	99.02
	LR	98.94	100.00	98.93	48.59	99.87
	KNN	99.58	86.96	99.71	75.00	99.14
GPMCH_2021	RF	99.44	100.00	99.43	52.94	99.91
	MLP	99.44	88.89	99.50	53.33	99.91
	DT	99.36	100.00	99.36	50.00	99.75
	SGD	98.52	100.00	98.51	30.00	99.40
	LR	98.94	100.00	98.93	37.50	99.95
	KNN	99.65	88.89	99.72	66.67	99.92
NWCH	RF	99.99	100.00	99.99	24.62	100.00
	MLP	99.64	100.00	99.64	1.12	99.98
	DT	99.99	93.75	99.99	33.30	96.87
	SGD	99.97	93.75	99.97	10.87	99.96
	LR	97.06	100.00	97.06	0.14	100.00
	KNN	99.94	100.00	99.94	6.38	100.00