International Journal of Advanced Research in Computer Science and Software Engineering (IJARCSSE)
A cluster is a collection of data objects that are similar to one another within the same cluster and are dissimilar to the objects in other clusters. Since K-means is widely used for general clustering, its performance is a critical point. This performance depends highly on initial cluster centers since it may converge to numerous local minima .There are many variation of K-means algorithm such as Lloyd's K-means clustering algorithm, grid based K-means algorithm, hierarchical K-means algorithm, etc. This paper is going to review all existing K-means algorithms and to propose a new and efficient K-means algorithm for clustering.