Design Of Personalization Recommendation Service In Egovernment Website Of A Development Zone

With the popularization and application of Internet, more and more enterprises and government apparatus aware it clearly and begin to set up their own E-Government systems to keep pace with modern management. But they are often confronted many problems as follows: the users are usually confused with so much information that he can't find what he want quickly; the government apparatus don't know how to provide personalized service for users according to his interesting; and don't know how to find out the user's browse patterns to adjust their management strategies, so that they can optimize their network's architectures and improve their efficiency. Web mining is a useful solution. E-Government recommendation system based on Web mining can help users find their interest information as well as improve their network's efficiency.This paper introduces concept, contents, technology of Data mining firstly, and introduces its overview, classification, working flow, feathers as well as development trends. Then, we make the unitary design of E-Government recommendation system and personalization, like the data anticipation. The data source in E-Government is mainly server-end data (it can be divided into the journal file and the inquiry data), which can be used to record user's behavior. Except that, there are proxy server-end data and user register information etc. After that, this paper studies the arithmetic of E-Government recommendation system and personalization, it uses the connection rule and its improvement algorithm to realize.Finally, This article unified a E-government website of the development zone, and designed an E-Government recommendation system based on Web usage mining, we introduce the functions of all the individual modules and its principle, its algorithm, its algorithm improvement and so on, with these, we can discover the visit rule of the users; and we can sum up the online users to a same kind, and carry on the recommendation of Web page according to this visit rule. And the system can also adjust the recommendation collection real-time by tracking the current visit of users unceasingly, and provide personalized visit for the users.