An imperialist competitive algorithm based fuzzy clustering algorithm
Mohammad Teimouri, Esmaeil Mehdizadeh · 2013
Clustering is one of the useful methods in many scientific fields. Fuzzy c-means (FCM) is the widely known approach to this problem, but it is highly depends on the initial state and converges to local optimum solution. The imperialist competitive algorithm (ICA) is a evolutionary algorithm based on human's socio-political evolution for searching problem space to find a near optimal solution. In this paper, we present a hybrid data clustering algorithm based on FCM and ICA, called FICA, which can find better cluster partition. The simulation results obtained by using the new algorithm on several benchmark data sets compared with those obtained by FCM, Genetic Algorithm (GA) and Particle Swarm Optimization (PSO) algorithm demonstrate the better performance of the new algorithm.