Research on Improved Clustering Algorithm on Web Usage Mining Based on Scientific Analysis of Web Materials

Abstract:

Clustering analysis is an important method to research the Web user’s browsing behavior and identify the potential customers on Web usage mining. The traditional user clustering algorithms are not quite accurate. In this paper, we give two improved user clustering algorithms, which are based on the associated matrix of the user’s hits in the process of browsing website. To this matrix, an improved Hamming distance matrix is generated by defining the minimum norm or the generalized relative Hamming distance between any two vectors. Then, similar user clustering are obtained by setting the threshold value. At the last step of our algorithm, the clustering results are confirmed by defining the clustering’s Similar Index and setting sub-algorithm. Finally, the testing examples show that the new algorithms are more accurate than the old one, and the real log data presents that the improved algorithms are practical.

• Periodical: Applied Mechanics and Materials (Volumes 63-64)
• Pages: 863-867
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.63-64.863
• Online since: June 2011
• Authors: Bin Li, Jin Yang, Cai Ming Liu, Jian Dong Zhang, Yan Zhang
• Keywords: Generalized Hamming Distance, User Clustering Algorithm, Web Usage Mining

References

[1] J. Srivastava, etal, Web usage mining: Discovery and applications of usage patterns from web data, SIGKDD Explorations, Vol. 1-2( 2000), p. l2-23.

DOI: 10.1145/846183.846188

[2] B. Mobasher and R. Cooley, Creating adaptive Web sites through Usage-based clustering of URLs, Proc of the 1999 IEEE Knowledge and Data Engineering Exchange workshop, New York: IEEE Press (1999), pp.32-37.

DOI: 10.1109/kdex.1999.836525

[3] G. Paliouras, et al, Clustering the users of large web sites into communities, Proc of the 17th Int Conf on Machine Learning, San Mateo: Morgan Kaufmann (2000), p. 7l9-728.

[4] Y. L. Yang, X. D. Guan and J. Y. You, Mining the page Clustering Based on the Content of Web Pages and the Site Topology, Journal of Software, Vol. 13-3(2002), pp.467-469.

[5] Q. B. Song and J. Y. Shen, An Efficient And Multi-Purpose Algorithm For Mining Web Logs, Journal of Computer Research & Development, Vol. 38-3(2001), pp.328-333.

[6] D. Beeferman and A. Berger, Agglomerative Clustering of a SearchEngine Query Log, Proceedings of the 6 ACM SIGKDD International Conference. Boston: ACM Press (2000), p.407-4l5.

DOI: 10.1145/347090.347176

[7] X. Y. Li and J. S. Yuan, Efficient Clustering Algorithm Used for Web Search, Computer Engineering, Vol. 32-20( 2006), pp.38-39.

[8] Y. Fu, K. Sandhu and M. Shih, A generalization-based approach to clustering of Web usage session, , Web Usage Analysis and User Profiling. New York: Springer-Verlag(2000), pp.21-38.

DOI: 10.1007/3-540-44934-5_2

[9] P. Kumar, P. R. Krishna and R. S. Bapi, et al, Rough Clustering of Sequential Data, Data & Knowledge Engineering, Vol. 63-2(2007), pp.183-199.

DOI: 10.1016/j.datak.2007.01.003