CirBiTree: Citrullination Site Inference Based on a Fuzzy Neural Network and Flexible Neural Tree

Table 9

The performances of different methods.

Length	Model	Sn (%)	Sp (%)	F1	MCC
17	Neural network	50.47	58.27	0.5252	0.0877
	Fuzzy neural network	62.17	60.17	0.6155	0.2234
	k Nearest neighbor	62.87	64.28	0.6332	0.2715
	Random forest	68.28	69.28	0.6862	0.3756
	Support vector machine	64.28	78.28	0.6912	0.4298
	Flexible neural tree	72.28	68.28	0.7086	0.4059
	CirBiTree	80.09	78.86	0.7960	0.5895
19	Neural network	52.81	62.82	0.5559	0.1571
	Fuzzy neural network	64.82	62.91	0.6421	0.2774
	k Nearest neighbor	60.28	68.28	0.6279	0.2865
	Random forest	70.21	71.28	0.7059	0.4149
	Support vector machine	71.82	72.28	0.7199	0.4410
	Flexible neural tree	75.92	74.82	0.7550	0.5074
	CirBiTree	81.01	80.09	0.8064	0.6110
21	Neural network	62.87	65.81	0.6381	0.2869
	Fuzzy neural network	65.28	61.82	0.6417	0.2712
	k Nearest neighbor	60.28	63.25	0.6119	0.2354
	Random forest	70.95	75.28	0.7252	0.4627
	Support vector machine	76.28	74.82	0.7573	0.5111
	Flexible neural tree	79.28	75.28	0.7773	0.5460
	CirBiTree	83.07	80.50	0.8202	0.6359