| With the development of society, there has been a growing need for lifelong learning and learn style for individuals, thus learning from the internet has become the main pattern. However, the present system of education on the web leaves much to be desired. It emphasize too much on the content while ignoring the difference of individuals. In this paper, we want to show you a quite fresh idea about an education system on the web to deal with some of the problem raised above.This paper has conducted an elaborate research on application of the technology from two perspectives, Web log mining and Web content mining. Combining their characteristics and relationships with the structure of Websites, we make some progress on the current methods concerned with Web mining and analyze a method of Web pages recommendation based on the Web content mining, and present some novel strategies integrating organically Web log mining and Web content mining that can work much more positively to enhance learning interests and efficiency of users by means of validly seeking for the user profile and raising the accuracy of recommendation. The main body of the essay would be as follows:(1) The paper summarizes and explores the classic methods and procedures of data preprocess for Web usage mining.(2) In the aspect of content mining, the paper facilitates the mined interests of users to be comprehended by transforming browsing behaves into the characterized vectors serving special vocabulary as the elements in the way of analyzing Web texts.(3) It also analyzes a recommendation algorithm based on pages corresponding hyperlinks with the Vector Space Model, and then analyzes another recommendation method based on collaborative filtering.(4) The paper applies the relevant theories and algorithms to the practice of designing the long-distance teaching system, creates system model, introduces the individualized processing engine which is the core part of system, and describes the structures of system functions. |
PDF Full Text Request