Web is a current powerful platform for discovering knowledge and to gain of needful information by analysing the Web data. Web mining, as a part of data mining includes various types and one among them is Web usage mining. Web usage mining is the application of data mining to discover required and interesting patterns from Web data which gives clear understanding of Web based applications. Web usage mining process starts with pre-processing of raw data, followed by clustering of data and finally visualization of the formed clusters effectively. Clustering is a technique which groups the complete data set into clusters such that data objects in each cluster will have some similarity. K-means is one such well known and most used clustering method. Usually K-means is used to cluster very large set of data. In K-Means algorithm, we need to calculate the distance between each of data object and cluster centre in every iteration. As K-means is partitioning and iterative method the clusters formed will be independent and compact. There will be random selection of K centres in the first iteration and obtain clusters by