E-commerce Web-based Data Mining Technology

With the popularization of Internet and the development of E-commerce, E-business Web sites are faced with more and more fierce competition. E-commerce provides abundant data resources for data mining, thus more and more business organizations are interested in how to make use of E-commerce information and mine the user interest pattern to understand user behavior, improve the usability of web site and provide personalized service for customers.Web mining is the traditional data mining technology used in Web, which can extract interesting and potentially useful patterns and implicit information from web documents or activity related to the World-Wide-Web. Thus studying this task has a definite application value and significance.The thesis discusses the character of data in E-commerce, expatiates the conception and technologies in detail, analyze the disadvantages of the model of Data Mining based on the Web log and construct a model of Data Mining based on Session, then it presents a Personalized recommender system based on similar model clustering algorithm. The system can capture similar model that exhibits a coherent pattern on either all or only a subset of dimensions by catching customer's visiting page sequence and constructing the top selling customer's behavior matrix model which provides some useful service information and scientific predicting data for business organizations.The designing of the system is available by simulating testing. Which gains some important and applicable information by mining E-commerce or business site.