Published Paper Details:

In an era where the use of technology has evolved manifolds, this resulted in the availability of information globally. Therefore, due to the multiple access of data over the internet, retrieval of desirable information has become a tedious and a time consuming task. Hence, use of various Data Mining techniques has become necessary for all the websites to allow the needed information to be accessed by the user. This results in the concept of Website Personalization, for which various Data Mining techniques are used, which further means to customize the website as per the need of almost every user visiting the website. One of the commonly used technique in Website Personalization is Clustering. Clustering is an unsupervised learning technique and is the process of grouping similar objects into different groups based on their centroid distance in a data set. Two most widely used techniques of clustering are K-Means and K-Medoids. In this paper, the concept of clustering is studied and the performance of both the techniques is analyzed graphically. The paper analyzes the performance using the methods i.e. Average Within Centroid Distance and Davies Bouldin on the Server Log Data.