Improving Recommendation Lists Through Topic Diversification
C. Ziegler, S. McNee, J. Konstan, and G. Lausen. Proceedings of the 14th International World Wide Web Conference, Chiba, Japan, ACM Press, (May 2005)
In this work we present topic diversification, a novel method designed to balance and diversify personalized recommenda- tion lists in order to reflect the user�s complete spectrum of interests. Though being detrimental to average accuracy, we show that our method improves user satisfaction with rec- ommendation lists, in particular for lists generated using the common item-based collaborative filtering algorithm. Our work builds upon prior research on recommender sys- tems, looking at properties of recommendation lists as en- tities in their own right rather than specifically focusing on the accuracy of individual recommendations. We introduce the intra-list similarity metric to assess the topical diver- sity of recommendation lists and the topic diversification approach for decreasing the intra-list similarity. We evalu- ate our method using book recommendation data, including online analysis on 361, 349 ratings and an online study in- volving more than 2, 100 subjects.
