Study of K-Means and Enhanced K-Means Clustering Algorithm

Abstract

Data clustering is a process of arranging data into groups or a technique for classifying a mountain of information into some manageable meaningful piles. The goal of clustering is to partitions a dataset into several groups such that the similarity within a group is better than among groups. K- means is one of the basic clustering algorithm which is commonly used in several applications, but it is computationally time consuming and the quality of the resulting clusters heavily depends on the selection of initial centroids.We can removed first limitation using the Enhanced K- Means algorithm. This paper represent the comparison analysis of basic K-Means clustering algorithm and Enhanced K- Means clustering algorithm which shows Enhanced K-Means algorithm more effective and efficient than Basic K-means algorithm.

Keywords: Basic K-Means clustering, Clustering, computational time complexity, centroids, Enhanced K-Means algorithm.