Applying and comparing hidden Markov model and fuzzy clustering algorithms to Web usage data for recommender systems

Authors

Shaghayegh Sahebi, University of Tehran, Iran
Farhad Oroumchian, University of Wollongong in DubaiFollow
Ramtin Khosravi, University of Tehran

RIS ID

27722

Publication Details

Sahebi, S., Oroumchian, F. & Khosravi, R. 2008, 'Applying and comparing hidden Markov model and fuzzy clustering algorithms to Web usage data for recommender systems', IADIS European Conference on Data Mining, 2008. Proceedings of, IADIS Press, Amsterdam, Netherlands, pp. 179-181.

Abstract

In this study, we apply and compare some of the methods of usage pattern discovery, like simple k-means clustering algorithm, fuzzy relational subtractive clustering algorithm, fuzzy mean field annealing (MFA) clustering and Hidden Markov Model (HMM), for recommender systems. We use metrics like prediction strength, hit ratio, precision, prediction ability and F-Score to compare the applied methods on the Web usage data. Fuzzy MFA and HMM acted better than other methods due to fuzzy nation of human behavior in navigation and extra information utilized in sequence analysis.