Improving Recommendation Based on Features’ Co-occurrence Effects in Collaborative Tagging Systems
Currently, recommender system becomes more and more important and challenging, as users demand higher recommendation quality. Collaborative tagging systems allow users to annotate resources with their own tags which can reflect users’ attitude on these resources and some attributes of resources. Based on our observation, we notice that there is co-occurrence effect of features, which may cause the change of user’s favor on resources. Current recommendation methods do not take it into consideration. In this paper, we propose an assistant and enhanced method to improve the performance of other methods by combining co-occurrence effect of features in collaborative tagging environment.
KeywordsRoot Mean Square Error Recommender System Vector Space Model Mean Absolute Error Recommendation Method
Unable to display preview. Download preview PDF.
- 1.A Survey of E-Commerce Recommender Systems (June 2007)Google Scholar
- 3.Bennett, J., Lanning, S., Netflix, N.: The netflix prize. In: KDD Cup and Workshop in conjunction with KDD (2007)Google Scholar
- 4.Cai, Y., Leung, H.F., Li, Q., Tang, J., Li, J.: Tyco: Towards typicality-based collaborative filtering recommendation. In: ICTAI (2), pp. 97–104. IEEE Computer Society (2010)Google Scholar
- 6.Herlocker, J., Konstan, J., Borchers, A., Riedl, J.: An algorithmic framework for performing collaborative filtering. In: Proceedings of the 22nd ACM SIGIR Conference, August 15-19, pp. 230–237 (1999)Google Scholar
- 8.Melville, P., Mooney, R.J., Nagarajan, R.: Content-boosted collaborative filtering for improved recommendations. In: Proceedings of the Eighteenth National Conference on Artificial Intelligence, pp. 187–192 (2002)Google Scholar