An Alternative Technique of Selecting the Initial Cluster Centers in the k-means Algorithm for Better Clustering


Although k-means works well in many cases it offers no accuracy guarantee and it has no idea to select ideal cluster representatives. This article presents a technique in which the initial cluster representatives in the standard k-means algorithm are chosen intelligently. Comparison of the quality of the clusters produced by the standard k-means algorithm… (More)


  • Presentations referencing similar topics