A computational strategy for information integration in meta-search agent systems

When several information search agents work together for a user's query, each will obtain a result. How to integrate these multiple results so as to provide the final result to the user has become an important issue in both distributed information retrieval and the application of multi-agent systems. In this paper, we present a computational strategy for information integration, which is based on ranking orders of multiple results, similarities of search agents, and reliabilities of results from different search agents. This strategy is essentially a framework that can be combined with any other method to form a powerful integration strategy in distributed information retrieval.