Research On Agent Learning From Personalized Intelligent Information Query System

With the development of Internet, people use a large number of information resources through Internet. However, with the rapidly increasing information in Internet, it is a challenge to select suitable information from so many informaion resources. The current informaiton retrievel technologys just offer keywords-based searching, but ingore the semantic content of the keywords itself. They still have many limits and their capabilities need to be enhanced. The paper proposes a Web personalized intelligence information query system on the basis of agent and describes the realizing method and architecture of the system.Agent technology is a new distributed computing technology, Agent is a computing unit which can run persistently and autonomously under distributed environment, it has several fertures such as initiative, interoperablity, reactiveness and autonomy. Agent technology provides an effective approach to solve newly emerging distrbuted application problems, and in the mean while, it provides a reasonable conceptual model to study features of distributed computing system comprehensively and accurately.In this paper, The goal of our reserach is to enhance the quality of information gather and the efficiency of information search, to decrease the repeated search or transfer of Web information, to realize the information search of user' interest-Oriented, to incarnate and reflect the user' request of the individuation information service, by implanting the Intelligent Agent with the abilities of self-adapted learning, information filtrate and automatic search.Our information Agent learning system consists of three main function modules: user Agent, searching Agent, filtering Agent. Three modules are independent and collaborate with each other. They are integrated as a whole and realize better the intelligence and adaptability of the system.