A Feature Selection Method of the Island Algorithm Based on Gaussian Mutation

Table 5

Experimental results of each algorithm on 6 UCI datasets.

Datasets	Average classification accuracy			Average fitness value			Average number of features
	KIAGM	IAGMFS	NIAGM	KIAGM	IAGMFS	NIAGM	KIAGM	IAGMFS	NIAGM
Breast cancer	0.7925	0.8033	0.7743	0.1799	0.1735	0.1743	4	3	4
Spectfheart	0.8054	0.8074	0.7220	0.2870	0.1276	0.2880	16	11	11
Vehicle	0.7176	0.7671	0.5630	0.2700	0.2256	0.3515	10	7	8
Tic-tac-toe	0.9172	0.9388	0.7349	0.2599	0.2196	0.2104	4	3	4
Wine	0.9100	0.9200	0.8700	0.1332	0.1262	0.1785	5	3	3
Zoo	0.9566	0.8740	——	0.1946	0.1564	——	6	4	——