Research Of Intelligent Search Engine System Based On Multi-Agent

Internet has already become the main place to gain and exchange information at present and in the future. However, it is an extremely challenging work to find the satisfied information in a great deal of information in internet. Nowadays traditional information retrieval tools can solve the problem of information resources orientation to some extent, but they are still unable to meet the people's increasing personalized and intelligent demand of information service.Confronted with the inherent characteristics of internet information resources and the current situation of existing information retrieval systems, this thesis has analyzed and studied the domestic and foreign research trends of search engine technology. Combined with the newest achievement in the field of artificial intelligence-Agent technology, an internet information retrieval approach Multi-Agent Intelligent Search Engine System (MAISES) is proposed and realized in this thesis.This system stresses in personalized service, and utilizes "user individual model" mould method designed by author which forms the user individual software model to enhance personalized degree. Moreover this system adopts initiative search and meta search to improve the system efficiency. Author has obtained abundant practical experience in both multi-agent system design and association research of multi-agent in one system. Some practical innovation in the personalized service research has been achieved. In order to enhance the personalized and intelligent degree of the system, the component search engine selection algorithm and the result merging algorithm which can help meta search engine to execute information search are designed in this thesis.