Finding group shilling in recommendation system

ABSTRACT

In the age of information explosion, recommendation system has been proved effective to cope with information overload in e-commerce area. However, unscrupulous producers shill the systems in many ways to make profit, and it makes the system imprecise and unreliable in a long term. Among many shilling behaviors, a new form of attack, called group shilling, appears and does great harm to the system. Because group shilling users are now well organized and become more hidden among various normal users, it is hard to find them by traditional methods. However, these group shilling users are similar to some extent, for they both shill the target items. We bring out a similarity spreading algorithm to find these group shilling users and protect recommendation system from unfair ratings. In our algorithm, we try to find these cunning group shilling users through propagating similarities from items to users iteratively. The experiment shows our similarity spreading algorithm improves the precision of the system and provides the system a reliable protection.

REFERENCES

• 1: John Canny, Collaborative Filtering with Privacy, Proceedings of the 2002 IEEE Symposium on Security and Privacy, p.45, May 12-15, 2002
• 2: Chrysanthos Dellarocas, Immunizing online reputation reporting systems against unfair ratings and discriminatory behavior, Proceedings of the 2nd ACM conference on Electronic commerce, p.150-157, October 17-20, 2000, Minneapolis, Minnesota, United States [doi>10.1145/352871.352889]
• 3: J. Konstan and J. Riedl. Good ratings gone bad: study shows recommender systems can manipulate users' opinions. CHI 2003 Conference on Human Factors in Computing Systems. April 2003.
• 4: 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]
• 5: Paul Resnick , Neophytos Iacovou , Mitesh Suchak , Peter Bergstrom , John Riedl, GroupLens: an open architecture for collaborative filtering of netnews, Proceedings of the 1994 ACM conference on Computer supported cooperative work, p.175-186, October 22-26, 1994, Chapel Hill, North Carolina, United States [doi>10.1145/192844.192905]
• 6: Badrul Sarwar , George Karypis , Joseph Konstan , John Reidl, Item-based collaborative filtering recommendation algorithms, Proceedings of the 10th international conference on World Wide Web, p.285-295, May 01-05, 2001, Hong Kong, Hong Kong [doi>10.1145/371920.372071]
• 7: Shyong K. Lam , John Riedl, Shilling recommender systems for fun and profit, Proceedings of the 13th international conference on World Wide Web, May 17-20, 2004, New York, NY, USA [doi>10.1145/988672.988726]

INDEX TERMS

Primary Classification:
  H. Information Systems
  H.3 INFORMATION STORAGE AND RETRIEVAL
      H.3.5 On-line Information Services
          Subjects: Commercial services

Additional Classification:
  H. Information Systems
  H.3 INFORMATION STORAGE AND RETRIEVAL
      H.3.5 On-line Information Services
          Subjects: Web-based services; Data sharing

General Terms:
Algorithms, Experimentation

Keywords:
collaborative filtering, group shilling, recommendation system