An Improved K-means Clustering Algorithm with refined initial centroids
Madhu Yedla, Sandeep Malle, T M Srinivasa · 2010
A final Clustering result of the k-means clustering algorithm greatly depends upon the correctness of the initial centroids. Generally the initial centroids for the k-means clustering are chosen randomly so that the selected initial centroids may converges to numerous local minima, not the global optimum. In this paper a new initialization approach to find initial centroids for k-means clustering is proposed. According to our experimental results, the Improved k-means Clustering Algorithm has the more accuracy with less computational time comparatively Original k-means clustering algorithm.