Table 4.

Specific Performance of Four Methods on Dataset RPI2241.

RPI2241	Accuracy	Sensitivity	Specificity	Precision	MCC
BGFE	0.9130	0.8772	0.9660	0.8590	0.8335
Raw feature	0.6438	0.6525	0.6313	0.6565	0.2881
Stacked auto-encoder	0.9041	0.8895	0.9329	0.8747	0.8156
Stacked auto-encoder without fine tuning	0.6438	0.6517	0.6327	0.6551	0.2879

The boldface indicates this measure performance is the best among the compared methods for individual dataset.