An Efficient Algorithm for Improved Web Usage Mining

Clustering is a web mining technique, which is a demanding field of research in which its latent applications create their own special requirements. Clustering is a method of grouping similar data into data sets, called clusters. Cluster analysis is a primary technique in conventional data analysis and many clustering methods have been recognized which requires number of clusters to be precise in advance and is dependent on initial starting points. In this paper, we present a new algorithm to discover data clusters for numerical and nominal data. Apriori algorithm generates large number of candidate set that is not efficient for both data. KeywordsWeb Mining, Clustering, Decision Tree, Apriori algorithm