Automatic attribute clustering based on genetic algorithms

In this paper, a clustering method of attributes based on genetic algorithms is proposed for feature selection. It combines both the average accuracy of attribute substitution in clusters and the cluster balance as the fitness function. Experimental comparison with the k-means clustering approach and with all combinations of attributes also shows the effectiveness of the proposed approach. Besides, the attributes with missing values can also be easily replaced by other attributes in the same clusters. The proposed approach is thus more flexible than the previous feature-selection techniques.