Abstract:
Clustering algorithms are computationally intensive, particularly when they are used to analyze large amount of high dimensional data. Exploring faster algorithms for clustering is a vital and often encountered research problem. k-mean algorithm is a well known partition based clustering technique and so many variations of this basic algorithm are proposed by various researchers. In order to explore the strength and weaknesses an attempt has been made to compare some of the existing variations of k-mean algorithms using synthetic sets of high dimensional data as benchmark for evaluation and some criteria is also evolved for comparison of clustering algorithms.
Sanjay Garg and Ramesh Chandra Jain , 2006. Variations of k-mean Algorithm: A Study for High-Dimensional Large Data Sets. Information Technology Journal, 5: 1132-1135.
Subscribe Today 
