MrDNM: A Novel Mutual Information-Based Dendritic Neuron Model

Table 4

Average classification accuracy (%) obtained by 30 runs for all compared classifiers.

Dataset		Orig	RENN	FaLKNR	AdaBoost	MultiBoost		MrDNM
WBCD	Accuracy (%)	95.28	96.14	96.28	94.99	95.85	96.62	96.80
	Rank	6	4	3	7	5	2	1
BUPA	Accuracy (%)	71.59	71.88	71.01	71.88	71.59	71.59	72.66
	Rank	5	2.5	7	2.5	5	5	1
IONO	Accuracy (%)	91.17	86.61	86.61	91.17	91.74	89.23	90.73
	Rank	2.5	6.5	6.5	2.5	1	5	4
PIMA	Accuracy (%)	75.39	76.69	75.91	75.26	75.13	78.07	76.80
	Rank	5	3	4	6	7	1	2
VOTE	Accuracy (%)	94.71	94.71	96.55	94.48	94.48	95.95	96.57
	Rank	4.5	4.5	2	6.5	6.5	3	1
	A.Rank	4.6	4.1	4.5	4.9	4.9	3.2	1.8