Table 2. Comparison of the prediction performance by the proposed method and some state-of-the-art works on the yeast dataset.

Model	Features	Classifier	SN(%)	PPV(%)	ACC(%)	MCC(%)
Our method (1st dataset)	MLD	RF	94.34±0.49	98.91±0.33	94.72±0.43	85.99±0.89
Our method (2nd dataset)	MLD	RF	92.67±0.79	99.51±0.23	93.83±0.61	84.05±1.47
Guo’s work (1st dataset)	ACC	SVM	89.93±3.68	88.87±6.16	89.33±2.67	N/A
	AC	SVM	87.30±4.68	87.82±4.33	87.36±1.38	N/A
Zhou’s work (1st dataset)	LD	SVM	87.37±0.22	89.50±0.60	88.56±0.33	77.15±0.68
Yang’s work (1st dataset)	LD (Cod1)	KNN	75.81±1.20	74.75±1.23	75.08±1.13	N/A
	LD (Cod2)	KNN	76.77±0.69	82.17±1.35	80.04±1.06	N/A
	LD (Cod3)	KNN	78.14±0.90	81.86±0.99	80.41±0.47	N/A
	LD (Cod4)	KNN	81.03±1.74	90.24±1.34	86.15±1.17	N/A

Here, N/A means not available.