Exploring Social and Behavioral Contexts for Information Retrieval

This project addresses three technical topics in behavioral and social data modeling, emphasizing theory, algorithmic development, and applications in Web search and recommendation systems: 1) Using social behavior data to better profile entities that are involved in Web search and online social services by exploring collaborative filtering methodology that can handle multiple relations among multiple entities; 2) Explore social relations and behaviors for learning to rank using integrative probabilistic models to characterize user behaviors and learn ranking functions; 3) Modeling the dynamics of social behavior data with temporal dynamics to capture user behavior and social environment chanage.

• This project is supported by NSF grant IIS-1116886

• Current publications
  1. S.H. Yang, A. Smola, B. Long, Hongyuan Zha and Y. Chang. Friend or Frenemy? Predicting Signed Ties in Social Networks. SIGIR, 2012.code
  2. K. Zhou and Hongyuan Zha. Learning Binary Codes for Collaborative Filtering. SIGKDD, 2012. details
  3. Z. Lu, Hongyuan Zha, X. Yang, W. Lin and Z. Zheng. A New Algorithm for Inferring User Search Goals Using Feedback Sessions. TKDE, 2012.
  4. K. Zhou, X. Li and Hongyuan Zha. Collaborative Ranking: Improving the Relevance for Tail Queries. CIKM, 2012. (short paper)
  5. K. Zhou, J. Bai, Hongyuan Zha and G. Xue. Leveraging Auxiliary Data for Learning to rank. ACM TIST, 2012.