Research On The Mechanism Of Web Information Retrieval Based On Mobile Agent

With the web information retrieval technology's running to today, it has indubitably brought large convenience to the information user. However, with the development of network and web technology, the web environment has grown up increasingly, the traditional web information retrieval technology couldn't meet the demand of information user yet. It has shown itself a good many defects, especially under the distributed web environment, that is: The traditional web information retrieval technology takes two kinds of mode essentially which is client/server or code-on-demand mode, under these two kinds of mode, the web information retrieval takes too many network bandwidth, the network's flux is too heavy, and it relies on the network connections'stability excessively; It can't guarantee the information retrieval's efficiency; The information retrieval's intelligence degree is too low; It lacks of the self learning ability and so on. Traditional web information retrieval technology shows itself the above defects, so how to resolve these problems becomes a very imminency problem in the correlative areas, the researches of this thesis just expands based on this problem.Mobile agent technology is a new technology in the distribute artificial intelligence area, it is an application branch of the agent technology and springs up not for long. At present, the definition of mobile agent is not unitive, different application environment gives it a different definition. Synthetically, this thesis consider mobile agent a kind of agent which has the mobility, it can move from one computer to another independently, it fulfill the user's appointed assignment. The mobile agent technology's rise is not for long and is not very mature, but its application experience in many areas shows that mobile agent technology has a very powerful technical dominance and application foreground, its technical dominance is that: mobile agent technology can reduce the network load, save the network bandwidth; it sustains long-distance alternation; encapsulate network protocol; sustains asynchrony independent execution; it sustains not-on-line computation; it doesn't depend on the flat roof and has the dynamic adaptability; it is prone to distribute services and it can buildup the muscularity and fault-tolerant ability of the system; it can provide individual services and so on. These advantages of mobile agent technology make it have a very wide application foreground. It is also this technology's rise provides a new way for the problems of the web information retrieval technology in this thesis.The researches in this thesis expand by surrounding the problem how to solve the defects in the traditional web information retrieval technology, it has creativity to import the mobile agent technology in the DAI to the web information retrieval area, it tries to resolve the problems in the web information retrieval area by this new technology, this is a kind of trialed research and is an innovation of this thesis.After analyzing the defects of the traditional web information retrieval technology and introducing for the mobile agent technology, this thesis integrates these two technologies, and constructs a web information retrieval system model based on mobile agent, this is a production of the thesis. This system model has four layers, those are application layer, agent layer, system flat roof layer and network layer, each layer has its application respectively. In this process, the thesis has introduced the main module's structure, principle and function amply, including mobile agent server, retrieval mobile agent, service agent, web service agent and yellow page server. Above this foundation, the thesis expatiates the system's executing mechanism. Theoretic research must have a demonstrative research to support. Accordingly, this thesis gave a example which is a web information retrieval system based on IBM's Aglet platform, expatiated the realization of main module on the Aglet platform. At last, the thesis introduced several key issues in the system, it includes the agent's mobility, communication mechanism, the system's fault-tolerant mechanism, its security and so on.Make a comprehensive view for this thesis, this thesis imported the mobile agent technology to the web information retrieval area aim at its defects, gave a new resolved precept which is web information retrieval technology based on mobile agent. This new precept can resolve the problem in traditional web information retrieval technology and have a lot of technical dominance, such as dynamic executing, asynchronous computing, choosing route independently and parallel searching. Thereby, this precept's putting forward certainly will promote the mobile agent technology's rapid development in the web information retrieval technology, at the same time, it made this research have both academic significance and practical significance. The penman believes that this created research certainly will arouse the attention for it in the relative academic area, and will raise the research upsurge for it, this research will promote the web information retrieval technology's development rapidly and much better, make it undergo the essential changing.