Most information retrieval (IR) interfaces are designed for a single user working with a dedicated interface. We present a system in which the IR interface has been fully integrated into a collaborative context of discussion or debate relating to the query topic. By using a tangible user interface, we support multiple users interacting simultaneously to refine the query. Integration with more powerful back-end query processing is still in progress, but we have already been able to evaluate the prototype interface in a real context of use, and confirmed that it can improve relevance rankings compared to single-user dedicated search engines such as Google. Author Keywords Tangible Interfaces, Computer Supported Collaborative Argumentation, Probabilistic Information Retrieval. ACM Classification Keywords H.5.3 [Group and Organization Interfaces]: Asynchronous interaction, Collaborative computing. H.3.3 [Information Search and Retrieval]: Relevance feedback, Search process.
Alan F. Blackwell, Mark Stringer, Eleanor F. Toye,