Abstract
The open nature of collaborative recommender systems present a security problem. Attackers that cannot be readily distinguished from ordinary users may inject biased profiles, degrading the objectivity and accuracy of the system over time. The standard user-based collaborative filtering algorithm has been shown quite vulnerable to such attacks. In this paper, we examine relevance measures that complement neighbor similarity and their influence on algorithm robustness. In particular, we consider two techniques, significance weighting and trust weighting, that attempt to calculate the utility of a neighbor with respect to rating prediction. Such techniques have been used to improve prediction accuracy in collaborative filtering. We show that significance weighting, in particular, also results in improved robustness under profile injection attacks.
This work was supported in part by the National Science Foundation Cyber Trust program under Grant IIS-0430303.
This is a preview of subscription content, log in via an institution to check access.
Preview
Unable to display preview. Download preview PDF.
Similar content being viewed by others
References
Lam, S., Riedl, J.: Shilling recommender systems for fun and profit. In: Proceedings of the 13th International WWW Conference, New York (May 2004)
O’Mahony, M., Hurley, N., Kushmerick, N., Silvestre, G.: Collaborative recommendation: A robustness analysis. ACM Transactions on Internet Technology 4(4), 344–377 (2004)
Mobasher, B., Burke, R., Bhaumik, R., Williams, C.: Towards trustworthy recommender systems: An analysis of attack models and algorithm robustness. ACM Transactions on Internet Technology 7(4) (2007)
Herlocker, J., Konstan, J., Borchers, A., Riedl, J.: An algorithmic framework for performing collaborative filtering. In: Proceedings of the 22nd ACM Conference on Research and Development in Information Retrieval (SIGIR 1999), Berkeley, CA (August 1999)
O’Donovan, J., Smyth, B.: Trust in recommender systems. In: Proceedings of the 10th International Conference on Intelligent User Interfaces (EC 2004), pp. 167–174. ACM Press, New York (2005)
Mobasher, B., Burke, R., Sandvig, J.J.: 1388–1393. In: Proceedings of the 21st National Conference on Artificial Intelligence, pp. 1388–1393. AAAI Press, Stanford, California (2006)
Breese, J., Heckerman, D., Kadie, C.: Empirical analysis of predictive algorithms for collaborative filtering. In: Uncertainty in Artificial Intelligence. Proceedings of the Fourteenth Conference, New Orleans, LA, pp. 43–53. Morgan Kaufman, Seattle, Washington (1998)
Williams, C., Bhaumik, R., Burke, R., Mobasher, B.: The impact of attack profile classification on the robustness of collaborative recommendation. In: Proceedings of the 2006 WebKDD Workshop, held at ACM SIGKDD Conference on Data Mining and Knowledge Discovery (KDD 2006), Philadelphia (August 2006)
O’Donovan, J., Smyth, B.: Is trust robust?: An analysis of trust-based recommendation. In: Proceedings of the 5th ACM Conference on Electronic Commerce (EC 2004), pp. 101–108. ACM Press, New York (2006)
Herlocker, J., Konstan, J., Tervin, L.G., Riedl, J.: Evaluating collaborative filtering recommender systems. ACM Transactions on Information Systems 22(1), 5–53 (2004)
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 2007 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Sandvig, J.J., Mobasher, B., Burke, R. (2007). Impact of Relevance Measures on the Robustness and Accuracy of Collaborative Filtering. In: Psaila, G., Wagner, R. (eds) E-Commerce and Web Technologies. EC-Web 2007. Lecture Notes in Computer Science, vol 4655. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-540-74563-1_10
Download citation
DOI: https://doi.org/10.1007/978-3-540-74563-1_10
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-74562-4
Online ISBN: 978-3-540-74563-1
eBook Packages: Computer ScienceComputer Science (R0)