Data clustering using eDE, an enhanced differential evolution algorithm with fuzzy c-means technique

Abstract

Clustering is the way toward sorting out items into groups whose individuals are comparative somehow. It is a gathering of articles that are intelligent inside, yet unmistakably not at all like the items having a place with different groups. Clustering of data plays a major part in efficient customer segmentation, organization of documents, information retrieval, extraction of topics, classification, collaborative filtering, visualization, and indexing. In the area of information retrieval systems, evolutionary algorithms work in a robust and efficient manner for clustering. To overcome the problem of local maxima, various nature-inspired metaheuristic algorithms like particle swarm optimization, artificial bee colony, and firey algorithms are considered. In this work, a variant of a differential evolution algorithm named enhanced differential evolution (eDE) is created. eDE is incorporated with the fuzzy c-means technique to perform clustering of data.