Research And Design Of Search Engine Based On Agent

In recent years, as the appearance of computers and networks, the Internet information go into the exponential growth, Search engine has become essential tools for the Internet. Traditional search engines advantage is informative, timely updates without manual intervention, but with the exponential growth of information, Massive search technology is increasingly obvious drawbacks, the return of information overload, coupled with a large number of redundant or irrelevant information. We must search results on its own or multiple secondary screening, virtually reduced the accuracy of the search. We need to have new technology to solve these problems. The research based on the optimization of multi-Agent search several key issues, from the second generation of search by the current problems and to conduct search engine optimization further.By a deep analysis of Web crawling, file processing, Chinese words segmenting, index building, for some faults to waste some network resource owing to filtrating after downloading, it is designed the search engine of professional information. The system was improved on Web crawling, Chinese words segmenting and Index building sub-system in order to realize a function to estimate correlative rate between a web site and all of sub-system are able to work compatibly. As three are following:1) The paper analyzes and studies the characteristics and main framework of traditional search engine.2) The paper analyzes and studies the main kinds of Agents and their characteristics. According to the character of search engine and the factors of realization, choose the Multi-Agent model to realize the search engine system.3) Based on the Agent model and search engine framework, the paper improves some AI algorithms, ideas and proposes web page characteristics withdrawing technique, which greatly improves the search engine intelligence and individuation of search engine. All the technique include:through the marked tree and the layering mark symbols, increase the system ability to judge the core meaning of the web page; adopting some intelligent algorithms, such as the user's interest research, multi-user's interest cooperation, increase the intelligence of the user interface Agent and user's individuation, and bring the information "Push and Pull" idea into the search engine, and introduce a JMAS model. The JMAS technology upgrades the system. Since that, it ensures that some sub-system of Web crawling, which works in some different locations, can exchange resource safely and reliably. Some information gathered by sub-system of Web crawling can be stored in local, and tested its performance.