Feature Selection Method Based on Artificial Bee Colony Algorithm and Support Vector Machines for Medical Datasets Classification

Table 9

Classification accuracies obtained by our method and other classifiers for the liver disorders dataset.

Author (year)	Method	Classification accuracy (%)
Lee and Mangasarian (2001) [7]	SSVM (10-fold CV)	70.33
van Gestel et al. (2002) [8]	SVM with GP (10-fold CV)	69.7
Gonçalves et al. (2006) [9]	HNFB-1 method	73.33
Özşen and Güneş (2008) [10]	AWAIS (10-fold CV)	70.17
	AIS with hybrid similarity measure (10-fold CV)	60.57
	AIS with Manhattan distance (10-fold CV)	60.21
	AIS with Euclidean distance (10-fold CV)	60.00
Li et al. (2011) [11]	A fuzzy-based nonlinear transformation method + SVM	70.85
Chen et al. (2012) [12]	(PSO) + 1-NN method (5-fold CV)	68.99
Chang et al. (2012) [13]	CBR + PSO (train: 75%-test: 25%)	76.81
Our study	ABCFS + SVM (train: 75%-test: 25%)	82.55
	ABCFS + SVM (10-fold CV)	74.81