Towards Improving Ant-Based Clustering - An Chaotic Ant Clustering Algorithm

Ant-based clustering as a nature-inspired heuristic algorithm has been applied in a data-mining context to perform both clustering and topographic mapping. It is derived from a basic model of behavior observed in real ant colonies. Early works demonstrated some promising characteristics of the ant-based clustering, but they did not extend to improve its performance, stability, convergence, and other key features. In this paper, we describe an improved version, called CACAS, adopting an important strategy of using chaotic perturbation to improve individual quality and utilized chaos perturbation to avoid the search being trapped in local optimum. We compare its performance with the K-means approach and ant-based clustering by evaluation functions and topographic mapping using a set of analytical data. Our results demonstrate CACAS is a robust and viable approach.