Clustering Analysis Based Learning of Web Mining

Aparna Upadhyay, Ravindra Gupta

Abstract: The World Wide Web has a giant amount of different forms of data and mining the data leads to knowledge discovery which is used in various fields. These discoveries need a proper way to be analysed for further use such as in machine learning, artificial intelligence etc. Clustering is a conventional method of analysing web data and giving best solutions by different evaluation methods. There are various clustering algorithms present but the accuracy and efficiency is what needed in analysis. In this paper, the comparisons of two of the major clustering algorithms i.e. k-means and Hierarchical algorithm is done and the best algorithm is shown through external evaluation method.

Keywords: Web Mining, K- means algorithm, Hierarchical algorithm, Euclidean distance function, Precision and Recall.