Data clustering using hierarchical deterministic annealing and higher order statistics

In this brief, we propose an extension to the hierarchical deterministic an nealing (HDA) algorithm for clustering by incorporating additional features into the algorithm. To decide a split in a cluster; the interdependency am ong all the clusters is taken into account by using the entire data distrib ution. A general distortion measure derived from the higher order statistic s (HOS) of the data is used to analyze the phase transitions, Experimental results clearly demonstrate the improvement in the performance of the HDA a lgorithm when the interdependency among the clusters and the HOS of the dat a points are also utilized for the purpose of clustering.