Take a look at the new version of this page: [ beta version ]. Tell us what you think.

ABSTRACT

Recommender systems based on user feedback rank items by aggregating users' ratings in order to select those that are ranked highest. Ratings are usually aggregated using a weighted arithmetic mean. However, the mean is quite sensitive to outliers and biases, and thus may not be the most informative aggregate. We compare the accuracy and robustness of three different aggregators: the mean, median and mode. The results show that the median may often be a better choice than the mean, and can significantly improve recommendation accuracy and robustness in collaborative filtering systems.

REFERENCES

• 1: Laurent Candillier , Frank Meyer , Françoise Fessant, Designing Specific Weighted Similarity Measures to Improve Collaborative Filtering Systems, Proceedings of the 8th industrial conference on Advances in Data Mining: Medical Applications, E-Commerce, Marketing, and Theoretical Aspects, July 16-18, 2008, Leipzig, Germany [doi>10.1007/978-3-540-70720-2_19]
• 2: F. Garcin, B. Faltings, and R. Jurca. Aggregating reputation feedback. In ICORE 09, 2009.
• 3: Jonathan L. Herlocker , Joseph A. Konstan , Loren G. Terveen , John T. Riedl, Evaluating collaborative filtering recommender systems, ACM Transactions on Information Systems (TOIS), v.22 n.1, p.5-53, January 2004 [doi>10.1145/963770.963772]
• 4: R. Jurca, F. Garcin, A. Talwar, and B. Faltings. Reporting incentives and biases in online review forums. ACM Transaction on the Web. to appear.
• 5: Bhaskar Mehta , Thomas Hofmann , Wolfgang Nejdl, Robust collaborative filtering, Proceedings of the 2007 ACM conference on Recommender systems, October 19-20, 2007, Minneapolis, MN, USA [doi>10.1145/1297231.1297240]
• 6: Bamshad Mobasher , Robin Burke , Runa Bhaumik , Chad Williams, Toward trustworthy recommender systems: An analysis of attack models and algorithm robustness, ACM Transactions on Internet Technology (TOIT), v.7 n.4, p.23-es, October 2007 [doi>10.1145/1278366.1278372]
• 7: Michael O'Mahony , Neil Hurley , Nicholas Kushmerick , Guénolé Silvestre, Collaborative recommendation: A robustness analysis, ACM Transactions on Internet Technology (TOIT), v.4 n.4, p.344-377, November 2004 [doi>10.1145/1031114.1031116]
• 8: Paul Resnick , Rahul Sami, The influence limiter: provably manipulation-resistant recommender systems, Proceedings of the 2007 ACM conference on Recommender systems, October 19-20, 2007, Minneapolis, MN, USA [doi>10.1145/1297231.1297236]
• 9: Paul Resnick , Rahul Sami, The information cost of manipulation-resistance in recommender systems, Proceedings of the 2008 ACM conference on Recommender systems, October 23-25, 2008, Lausanne, Switzerland [doi>10.1145/1454008.1454033]
• 10: Kevin Walsh , Emin Gün Sirer, Fighting peer-to-peer SPAM and decoys with object reputation, Proceedings of the 2005 ACM SIGCOMM workshop on Economics of peer-to-peer systems, August 22-22, 2005, Philadelphia, Pennsylvania, USA [doi>10.1145/1080192.1080204]

INDEX TERMS

Primary Classification:
  H. Information Systems
  H.3 INFORMATION STORAGE AND RETRIEVAL
      H.3.3 Information Search and Retrieval

General Terms:
Performance, Reliability, Security

Keywords:
accuracy, recommender systems, robustness