Clustering of Web Usage Data Using Enhanced K-means Clustering Algorithm

Clustering is the process of partitioning orgrouping a given set of patterns into disjointclusters. Clustering web usage data requiresdeveloping specialized techniques based on theweb log data. The methodology has to improve thedata preprocessing as well as the quality of theclusters. Traditional K-means algorithm is widelyused clustering algorithm with wide range ofapplication. Since K-means algorithm has somedisadvantages. The proposed system intends toimplement an enhanced K-means clusteringalgorithm to get more accurate and effectivecluster results. The algorithm will carry out intwo steps which include finding initial centroidsand clustering stage. The testing of the systemuses web log data from NASA web usage data forclustering. The performance and accuracy of thesystem are also measured