论文信息 - Mining Interest Association Rules in Website Based on Hidden Markov Model

Mining Interest Association Rules in Website Based on Hidden Markov Model

Generally, a user will access a Web site with a certain interest. Mining Web users' interest access patterns has been an important research direction in Web usage mining. These patterns are a kind of the special interest association rules essentially. In this paper, we propose a new approach for mining such rules based on hidden Markov model (HMM). In our approach, pages' contents and Web server's log need to be preprocessed firstly. Next we present some definitions of users' access interest in a Web site. In addition, a new incremental algorithm Hmm_R is given to discover the interest association rules. Finally, we report on experiments conducted with simulative and real data and then testify that the algorithm can find all interest association rules efficiently.

Zhiguo Zhu | Guishi Deng

[1] Lawrence R. Rabiner,et al. A tutorial on hidden Markov models and selected applications in speech recognition , 1989, Proc. IEEE.

[2] Jaideep Srivastava,et al. Web usage mining: discovery and applications of usage patterns from Web data , 2000, SKDD.

[3] Michael Kaufmann,et al. Usage-based visualization of web localities , 2001 .

[4] Pattie Maes,et al. Footprints: History-Rich Web Browsing , 1997, RIAO.

[5] Mark Levene,et al. Data Mining of User Navigation Patterns , 1999, WEBKDD.

[6] Pedro M. Domingos,et al. A machine learning approach to web personalization , 2002 .

[7] Oren Etzioni,et al. Adaptive Web Sites: Conceptual Cluster Mining , 1999, IJCAI.

[8] Bamshad Mobasher,et al. Web Usage Mining and Personalization , 2004, The Practical Handbook of Internet Computing.

[9] Jaideep Srivastava,et al. Grouping Web page references into transactions for mining World Wide Web browsing patterns , 1997, Proceedings 1997 IEEE Knowledge and Data Engineering Exchange Workshop.