A hybrid clustering algorithm based on kmeans and ant lion optimization

Abstract

Kmeans is one of the most-efficient hard-clustering algorithms. It has been successfully applied to a number of problems. However, the efficiency of kmeans is dependent on its initialization of cluster centres. Different swarm intelligence techniques are applied for clustering problem. In this work, we have considered Ant Lion Optimization (ALO) which is a stochastic global optimization models. In this work Kmeans has been integrated with ALO for optimal clustering. The statistical measures of different performance metrics has been calculated and compared. The proposed method performs preferably better than Kmeans and PSO-Kmeans in terms of sum of intracluster distances and F-measure.