Table 3.

Specific Performance of Four Methods on Dataset RPI1807.

RPI1807	Accuracy	Sensitivity	Specificity	Precision	MCC
BGFE	0.9600	0.9344	0.9989	0.9117	0.9217
Raw feature	0.9349	0.9508	0.9308	0.9400	0.8688
Stacked auto-encoder	0.9396	0.9029	0.9994	0.8651	0.8830
Stacked auto-encoder without fine tuning	0.9645	0.9672	0.9688	0.9590	0.9281

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