Table 6.

Performance comparison of existing approaches on the independent test dataset (BID18).

Methods	Classifier		SPE	SEN	PRE	ACC	F1	MCC
HEP	SVM		0.76	0.6	0.84	0.79	0.70	0.56
PredHS-SVM	SVM		0.93	0.79	0.59	0.83	0.68	0.57
iPPHOT	SVM		0.586	0.462	0.794	0.650	0.584	0.353
KFC2a	SVM		0.73	0.55	0.74	0.73	0.63	0.44
KFC2b	SVM		0.87	0.64	0.55	0.77	0.60	0.44
PCRPi	Bayesian network		0.75	0.51	0.39	0.69	0.44	0.25
MINERVA	SVM		0.90	0.65	0.44	0.76	0.52	0.38
APIS	SVM		0.76	0.57	0.72	0.75	0.64	0.45
KFC	Decision trees		0.85	0.48	0.31	0.69	0.38	0.19
Robetta	Knowledge-based method		0.88	0.52	0.33	0.72	0.41	0.25
FOLDEF	Knowledge-based method		0.88	0.48	0.26	0.69	0.34	0.17