An intelligent metasearch engine for the World Wide Web

Machine learning and information retrieval techniques are applied to metasearch on the World Wide Web as a means of providing user specific relevant documents in response to user queries. A metasearch agent works in conjunction with a user to provide daily sets of relevant documents. Users provide relevance feedback which is incorporated into future results by a choice of machine learning algorithms.; Using a fixed ranking method, the algorithms incorporating relevance feedback perform much better than those that do not. Furthermore, using heterogeneous information sources on the World Wide Web is shown to be effective in short and long term usage.