Enhancement of K-Means algorithm using ACO as an optimization technique on high dimensional data

Clustering is a distribution of data into groups of similar objects such that the objects in a group will be similar (or related) to one another and different from (or unrelated to) the objects in other groups. The concept of clustering applications is particularly in the context of information retrieval and in organizing web resources. The objective of clustering is to find out information and in the present day context, to locate most relevant resources. In data mining, K-Means clustering is a method of cluster analysis which aims to partition n observations into k clusters in which each observation belongs to the cluster with the nearest mean. Though the K-Means is one of the best clustering algorithms, the quality is based on the starting condition and it may converge to local minima. There is not much of work done by the researchers to improve the cluster quality after grouping. We have proposed a novel method to improve the cluster quality on high dimensional data set by ant based refinement algorithm. The Ant Colony Optimization algorithm (ACO) is one of the most widely used probabilistic technique for solving computational problems which can be reduced to finding good paths through graphs. An ant is a simple computational agent in the ant colony optimization algorithm. It develops an iterative solution for any problem at hand. The intermediate solutions can be used to arrive at the final solution. The proposed algorithm is tested using data from different domain and the results show that refined initial starting points and post processing refinement of clusters based on ACO can lead to improved solutions in terms of entropy, time taken and accuracy of clusters.