Distributed information search with adaptive meta-search engines

Abstract

With the flourishing development of E-Commerce and the exponential growth of information produced by businesses, organizations and individuals on the Internet, it becomes more and more difficult for a person to find information efficiently by only using one or a few search engine(s), due to the great diversity among heterogeneous sources. This paper proposes a method for building adaptive meta-search engines with which (1) users can express their information needs sufficiently and easily, and (2) those sources that may answer user queries best can be selected, and the mapping between user queries and the query capabilities of target sources is performed more accurately. Experiments show that this adaptive method for constructing a meta-search engine can achieve more precision than traditional ways. The method can be applied to all kinds of information integration systems on the Internet, as well as on corporate Intranets.