A comparative analysis of results data clustering with variants of differential evolution optimization

Differential evolution (DE) has emerged as one of the fast, robust, and efficient global search heuristics algorithm of current interest. This paper describes an application of DE to the clustering of unlabeled data sets. Here it has been used different variants of DE for clustering the data set. It is shown how different variants of DE can be used to find the centroids of a user specified number of clusters. These algorithms are evaluated on some real life datasets and compared their performances. The convergence characteristics of each variant are shown for different datasets. This study may be very useful for researchers for comparison purposes.