Research And Application Of Information Search Engine Technology Based On Multi-Agent

With the great development of WWW, we need a valid information tool to navigate facing great deal of information need in the WWW. Search engine provides information search function. Along with the development of multi-agent technology, search engine based on multi-agent also be researched extensively. This paper develops a testing system for the Search Engine Based on Multi- Agent(SEBMA) on the basis of the collection and process of the Internet information and the research of multi-agent. It realizes parallel and distributing process of the information collectioning course, better resolves the problem of the large-scale Internet information collection and the improvement of the accurate for information search by utilizing the framework on the ground of multi-agent.Firstly,the paper proposes the demand analysis of the testing system for the SEBMA,describes the partition of the system characteristics and the analysis of basic examples. Secondly,it explaines the whole structure of the SEBMA and the function of each agent-module,introduces in detail the structural method of plug-in agent ,the communicate language of agent and the management of the host agent. Thirdly, it analyzes the technology of the spiders,the mobile long-distance collecting agent, the Chinese participle and the whole-text index,the personal-interest and the information search. Finally,it puts forward the design and development and relating technology of the system and illuminates the system testing environment and testing cases and testing results.The aim is to develop a testing and verifying system , so the software has still not put to actual application. The paper presents the further development plan.