Research On Web Information Mining Based On Intelligent Agent

Web is a global net with various distributed sites and information resources over the world without a unified management system and structure, which result to the difficult information searching. This paper presents a detailed introduction of the current research situation and research significance home and abroad upon which the paper analyzed the key and difficulties in information mining and the traits or deficiencies in information mining research. An Agent-based Web IE system is designed on the basis of practical experiences and understanding of theories about Agent, data mining. The paper also illustrates the aims and the behaviors of each Agent. The division of each database and communicating methods of each Agent with basic test is also detailed analyzed in this paper.I. Detailed introduction of the traits and classification of Agent technology and analysis of advantages of Agent technology in data mining. The paper presents the analysis of various theories of Web mining including definition of Web mining, relationship between Web mining and Web information research, methods and objection of Web mining, etc.II. An Agent-based Web IE system is designed. The general objective of Web information mining is divided into different agents. The function of each Agent and the key technology in implement of each Agent is analyzed.III. Elaboration compostion about each knowledge base and database on all systems, the environment and Agent function can be achieved in the system development.IV. An XML-based Database Mining Agent on Web pages was developed using Java programming language.Because of combining the domain-specific semantic feature of the interested information and the Web page format feature in the definition of extraction rules, the proposed system obtains better reusability and adaptability. Moreover, The ability that agent can perceive how the Web site update and further more adapt the rules initiatively also contributes to enhancing the adaptability on some extent. In addition, the semi-automatic learning method in which the user participates simplifies the learning process.