A Vector Space Equalization Scheme for a Concept-based Collaborative Information Retrieval System

May 2003

Proceedings of the Sixteenth International Florida Artificial Intelligence Research Society Conference (FLAIRS 2003)

Volume

Proceedings of the Sixteenth International Florida Artificial Intelligence Research Society Conference (FLAIRS 2003)

All Papers

This paper describes a vector space equalization scheme for a concept-based collaborative information retrieval system; evaluation results are given. The authors previously proposed a peer-to-peer information exchange system that aims at smooth knowledge and information management to activate organizations and communities. One problem with the system arises when information is retrieved from another’s personal repository since the framework’s retrieval criteria are strongly personalized. The system is assumed to employ a vector space model and a concept-base as its information retrieval mechanism. The vector space of one system is very different from that of another system, so retrieval results would not reflect the requester’s intention. This paper presents a vector space equalization scheme, the automated relevance feedback scheme, that compensates the differences in the vector spaces of the personal repositories. A system that implements the scheme is realized and evaluated using documents on the Internet. This paper presents implementation details, the evaluation procedure, and evaluation results.