Framework, algorithms and experiments in Social Information Retrieval (SIR) incorporating models of trust and reputation and notions authoritativeness and popularity, and their relations to relevance.
Recently the fields of information retrieval and social network analysis have been bridged by SIR models, which extend conventional IR to incorporate the social context of search and recommendation. A particular aspect of SIR is the modeling of the trust and reputation of the members of the social network, which is then typically used for enhancing recommendation systems.
Our goal in this project is to understand the role of trust and reputation in a search scenario and to develop a comprehensive SIR framework, consisting of hybrid social and data networks, where relevance is modeled in terms of the trust propagated through the network. Our current research focus is on trust propagation mechanisms that differentiate between authoritative and popular approvals of an information item.
- Our paper, presented at CIKM 2008, won Best poster prize.
- Gabriella Kazai and Natasa Milic-Frayling, Trust, authority and popularity in social information retrieval, in CIKM '08: Proceeding of the 17th ACM conference on Information and knowledge mining, ACM, New York, NY, USA, 2008.