Data clustering using improved fire fly algorithm

Abstract

Clustering is considered as one of the most important techniques for data mining that is used for data analysis in some areas such as text identification, image processing, economic science, and spatial data analysis. Several algorithms have been proposed for solving the problem clustering. These algorithms are using different techniques. Firefly algorithm was inspired by the process of producing twinkle lights of this insect, and is considered as one of the designed base on a collective behavior of insects. In this paper, the evolutionary algorithm of Firefly is used for solving clustering problem. The proposed algorithm is compared with firefly algorithm, differential evolution algorithm and k-means algorithm on some important data sets from database UCI. According to the results, this algorithm is more appropriate for better clustering rather than firefly algorithm.