A New Ant-based Clustering Algorithm on High Dimensional Data Space

Abstract

Ant-based clustering due to its flexibility, stigmergic and self-organization has been applied in a variety areas from problems arising in commerce, to circuit design, and to text-mining, etc. A new ant-based clustering method named AMC algorithm has been presented in this paper. Firstly, an artificial ant movement(AM) model is presented; secondly, the new ant clustering algorithm has been constructed based on AM model. In this algorithm, each ant is treated as an agent to represent a data object, each ant has two states: resting state and moving state. The ant’s state is controlled by two predefined functions. By moving dynamically, the ants form different subgroups adaptively, and consequently the whole ant group dynamically self-organized into distinctive and independent subgroups within which highly similar ants are closely connected. This algorithm can be accelerated by the use of a global memory bank, increasing radius of perception and density-based ‘look ahead’ method for each agent. Experimental results show that the AMC algorithm is much superior to other ant clustering methods. It is adaptive, robust and efficient, and achieves high autonomy, simplicity and efficiency. It is suitable for solving high dimensional and complicated clustering problems.