Collection Profiling for Collection Fusion in Distributed Information Retrieval Systems

Discovering resource descriptions and merging results obtained from remote search engines are two key issues in distributed information retrieval studies. In uncooperative environments, query-based sampling and normalizing scores based merging strategies are well-known approaches to solve such problems. However, such approaches only consider the content of the remote database and do not consider the retrieval performance. In this paper, we address the problem that in peer to peer information systems and argue that the performance of search engine should also be considered. We also proposed a collection profiling strategy which can discover not only collection content but also retrieval performance. Web-based query classification and two collection fusion approaches based on the collection profiling are also introduced in this paper. Our experiments show that our merging strategies are effective in merging results on uncooperative environment.