Ant clustering algorithm based on point symmetry

This paper proposed a new point symmetry-based ant clustering algorithm which can defect the number of clusters and the proper partitions from data sets when data sets possess the property of symmetry. In the proposed algorithm, a revised ant clustering algorithm is presented which can reduce the running time of standard ant clustering algorithm. Each ant represents a data object. It will decide its next moving position according to similarity function and probability converting function between it and its neighbors. At the same time it will update its cluster number according to clustering rules. Each ant only depends on a little local information to cluster. Assignment of points to different clusters is done based on point symmetry distance rather than the traditional Euclidean distance. Kd-tree-based nearest neighbor search is used to reduce the complexity of computing PS-based distance. The effectiveness of point symmetry-based ant clustering compared to standard ant clustering is demonstrated for one artificial and one real-life data sets.