Supporting social recommendations with activity-balanced clustering

ABSTRACT

In support of social interaction and information sharing, online communities commonly provide interfaces for users to form or interact with groups. For example, a user of the social music recommendation site last.fm might join the "First Wave Punk" group to discuss his or her favorite band (The Clash) and listen to playlists generated by fellow fans. Clustering techniques provide the potential to automatically discover groups of users who appear to share interests. We explore this idea by describing algorithms for clustering users of an online community and automatically describing the resulting user groups. We designed these techniques for use in an online recommendation system with no pre-existing group functionality, which led us to develop an "activity-balanced clustering" algorithm that considers both user activity and user interests in forming clusters.

REFERENCES

• 1: Arindam Banerjee , Joydeep Ghosh, Scalable Clustering Algorithms with Balancing Constraints, Data Mining and Knowledge Discovery, v.13 n.3, p.365-395, November 2006 [doi>10.1007/s10618-006-0040-z]
• 2: Burke, R. Integrating Knowledge-Based and Collaborative Filtering Recommender Systems. Workshop on Artificial Intelligence for Electronic Commerce, 1999.
• 3: Gale, D., Shapley, L. College Admissions and the Stability of Marriage. American Mathematical Monthly, 69(1), 1962.
• 4: F. Maxwell Harper , Dan Frankowski , Sara Drenner , Yuqing Ren , Sara Kiesler , Loren Terveen , Robert Kraut , John Riedl, Talk amongst yourselves: inviting users to participate in online conversations, Proceedings of the 12th international conference on Intelligent user interfaces, January 28-31, 2007, Honolulu, Hawaii, USA [doi>10.1145/1216295.1216313]
• 5: He, J., Tan, A., Tan, C., Sung, S. On Quantitative Evaluation of Clustering Systems. In Information Retrieval and Clustering, Kluwer Academic Publishers, 2002.
• 6: A. K. Jain , M. N. Murty , P. J. Flynn, Data clustering: a review, ACM Computing Surveys (CSUR), v.31 n.3, p.264-323, Sept. 1999 [doi>10.1145/331499.331504]
• 7: Popescul, A., Ungar, L. Automatic Labeling of Document Clusters, Unpublished Manuscript, Available at http://citeseer.nj.nec.com/popescul00automatic.html, 2000.
• 8: Jenny Preece, Online Communities: Designing Usability and Supporting Socialbilty, John Wiley & Sons, Inc., New York, NY, 2000
• 9: 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]
• 10: Shilad Sen , Shyong K. Lam , Al Mamunur Rashid , Dan Cosley , Dan Frankowski , Jeremy Osterhouse , F. Maxwell Harper , John Riedl, tagging, communities, vocabulary, evolution, Proceedings of the 2006 20th anniversary conference on Computer supported cooperative work, November 04-08, 2006, Banff, Alberta, Canada [doi>10.1145/1180875.1180904]

INDEX TERMS

Primary Classification:
  H. Information Systems
  H.5 INFORMATION INTERFACES AND PRESENTATION (I.7)
      H.5.3 Group and Organization Interfaces
          Subjects: Web-based interaction

Additional Classification:
  I. Computing Methodologies
  I.5 PATTERN RECOGNITION
      I.5.3 Clustering
          Subjects: Algorithms

General Terms:
Algorithms, Design, Human Factors

Keywords:
activity-balanced clustering, user group summarization