Study On Framework Of Web Data Mining System In Electronic-Commerce

With the development of network technology and popularization of Internet application, E-commerce is booming with surprising speed as a modern new business model and affecting the economic operation of the entire community structure. E-commerce system will produce a lot of important information of original data and access data everyday, these data is mainly storage in Web server with a Web log file format. How to convert these heterogeneous, uncertainty and unstructured data into useful information become one of the important tasks of the electronic commerce system constructionThe data mining technology is a process of extracting potential useful information and knowledge from a large number of complex data. The Web data mining technology used in the construction of e-commerce system has an important significance. At present, almost all of the major e-commerce systems are using a variety of forms of data mining techniques to varying degrees and they can provide customers with personalized service, for example Amazon, dangdang and so on. However, most systems only provide a simple statistical analysis based on OLAP and most systems are based on closed system architecture design, it is inadequate to adapt to the changing needs of e-commerce Mining Application. This topic is committed to building an e-commerce environment based on open data mining framework and using the Web Services and Agent technology applications to this architecture.This paper mainly made the following work:(1) Analyzed and summarized the current status of e-commerce system architecture and dynamic, Pointed out the problem that exist in traditional e-commerce Web mining system. Through the research of Web services and Mobile Agent technology-related theory, put forward the ideas that use the Web Services distributed ideology and Intelligent Agent to achieve the superiority of the Web data mining framework in e-commerce environment.(2) In the foundation of researching the variety of Web mining algorithms application in the e-commerce environment, improved the shortcomings that exist in the key Apriori association rules and the core of k-means clustering algorithm and tested its effectiveness for the improved algorithm, At the same time in the stage of mining architecture, package the variety of Web algorithms and do together to complete excavation task.(3) Research the realization technology in e-commerce System and compare the realization of J2EE platform-based implementation and based on other platform. At last, design a simple prototype system use the J2EE platform-based three-tier structure model and MVC design patterns.