The Research Of A New Information Search System Based On Mobile Agent

The 21st century is an era in which the human beings are making for the informational society, and in which network is the most important tool in daily life, and the construction of information-highway makes great progress. The environment of network computing is also changed from Client/Server via Client/Network to Client/Virtual Environment, hi this case, there are many models of mobile computing, such as the model of Client/Server, Code-On-Demand and so on. But these models have their own shortages, for example, in the process of communicating with remote host or remote server, most of them make use of synchronous communication, and the waiting time is very long; when used hi some task, these models work only by serial model.Mobile Agent, which has lots of advantages in distributive computing or mobile computing (for example, the use of Mobile Agent may decrease the amount of communication, and communicate by asynchronous model.), is a hybrid of Internet technology and Artificial Intelligence. Owing to characteristics of intelligence, Mobile Agent can decide the destination by himself, it is the very advantage that can balance the load and make the browsing and searching more efficient and faster. These characteristics of Mobile Agent promote it to be extensively applied in distributive computing environment, such as the management of network, electronic commerce, information search, etc. The model of Mobile Agent will become the most important, and fashionable way in network computing environment in future.This thesis constructs a model of architecture based on Agent theoretical research, such as Agent theoretical model and structure. This Mobile Agent System consists of four main parts: the service environment of Mobile Agent, Mobile Agent, dock server, and information service center. The thesis focuses on the model of communication and the migrating mechanism. Basing on the model of architecture, we design a based-on Mobile Agent information-searching environment. The traditional search-engine searches information online by a method based on Client/Server, and keeps gigantic information in server information warehouse; it also completely relies on network s performance, and it does not support transmitting information by asynchronous model. So we request a personalized intelligent information-searching system. Custom submits search information to Information Mobile Agent (IMA) through Personal Intelligent Agent (PIA), and then IMA moves to destination according to tr unerary scheme, where IMA communicates with Information Server Agent (ISA), and IS. egins to search information in information warehouse according to the searching-' form: on IMA brought. When the search is completed, ISA submits the consequence to IMA, an- IMA migrates the next host or goes back. Such an Agent society may provide more efficient 'formation services and has more extensive prospect.