Novel Clustering Algorithms Based on Improved Artificial Fish Swarm Algorithm

Abstract

An improved artificial fish swarm algorithm (IAFSA) is proposed, and its complexity is much less than the original algorithm (AFSA) because of a new proposed fish behavior. Based on IAFSA, two novel algorithms for data clustering are presented. One is the improved artificial fish swarm clustering (IAFSC) algorithm, the other is a hybrid fuzzy clustering algorithm that incorporates the fuzzy c-means (FCM) into the IAFSA. The performance of the proposed algorithms is compared with that of the particle swarm optimization (PSO), k-means and FCM respectively on Iris testing data. Simulation results show that the performance of the proposed algorithms is much better than that of the PSO, K-means and FCM. And the proposed hybrid fuzzy clustering algorithm avoids the FCM's weakness such as initialization value problem and local minimum problem.