Table 2.

Performance Comparison on Dataset I and Dataset II

	Dataset I						Dataset II
Methods	Accu	Sen	Spe	Pre	CC	F1	Accu	Sen	Spe	Pre	CC	F1
PredHS-SVM	0.88	0.75	0.93	0.79	0.69	0.76	0.87	0.86	0.87	0.84	0.74	0.84
PredHS-Ensemble	0.82	0.85	0.81	0.60	0.60	0.69	0.87	0.91	0.84	0.80	0.74	0.85
Robetta	0.80	0.51	0.9	0.62	0.44	0.56	0.77	0.51	0.96	0.89	0.54	0.65
FOLDEF	0.78	0.31	0.93	0.59	0.31	0.40	0.70	0.31	0.98	0.91	0.41	0.46
KFC	0.79	0.55	0.87	0.58	0.43	0.57	0.74	0.55	0.87	0.75	0.45	0.64
MINERVA2	0.84	0.58	0.93	0.72	0.55	0.64	0.81	0.58	0.97	0.93	0.62	0.72
HotPoint	0.72	0.54	0.78	0.44	0.29	0.48	0.73	0.54	0.86	0.73	0.43	0.62
APIS	0.76	0.43	0.87	0.54	0.33	0.46	0.75	0.74	0.75	0.70	0.50	0.71
KFC2a	0.82	0.55	0.91	0.66	0.49	0.59	0.83	0.84	0.82	0.78	0.67	0.80
KFC2b	0.78	0.47	0.88	0.60	0.38	0.51	0.81	0.74	0.86	0.82	0.62	0.75

Six performance measures are used: accuracy (Accu), sensitivity (Sen), specificity (Spe), precision (Pre), CC, and F1 score. The highest values are highlighted in bold.