Web Usage Mining Application In Distance Education

With the rapid development of network and communication, the distance education system based on WWW (World Wide Web) has developed progressively at home and abroad at present and a lot of universities have set up their own distance learning websites. But such WWW-based distance learning websites have two obvious shortcomings: the existing distance educational websites are basically all static; the existing distance educational websites can not educate the students in accordance with their aptitude and lack the distinction. While, these websites have accumulated a large amount of useful teaching information, which provides a solution to the above problems—the use of Web Data Mining with the combination of traditional Web Data Mining technology and Web technology.Web Data Mining can be categorized into three kinds, among which Web Usage Mining focuses mainly on the information on Web server, including contents such as log file, user's registration information, etc. Its results are usually the users' common behavior and interest, their personal visiting preferences, habits and modes etc. It facilitates the rational organization of network information and improvement of service quality and can be used for finding the systematic functional bottleneck, optimizing the website structure, improving systematic security, raising the validity of users' visits, finding users' need and interest and offering intelligent service, etc.Firstly, this essay has summarized the concept and function of Web Data Mining with the Web Data Mining classifications and the detailed use of Web Data Mining. After analyzing the log data source, it has discussed especially the preprocessing course of Web Data Mining, introduced the contents such as filter condition form, heuristic rule, maximal forward route, etc. and analyzed and summarized several algorithms in data clearing up, user recognition, conversation recognition, route supplementary and affairs recognition, etc. along with the explanation of the realizing course.Secondly, this essay has also discussed the related regular algorithm especially—to mine for the relevant web pages that appear together in conversation most frequently probably without hyperlinks to join each other directly between these web pages. As to its classical Apriori algorithm, on the basis of analyzing its principle and property, the essay enhances two joining and pruning steps with the obvious improvement on timeliness and validity.At the end of the essay, it also applies the basic theories of Web Data Mining to the distance education. After the analysis of systematic framework of Web Usage Mining used in distance education, it has suggested an application model and has explained each module's function. The advantage of this model lies in utilizing contents such as the log information, learners' information as users, etc. and receiving the interesting modes applied to the distance education system according to these modes to improve personalized service which can be more beneficial to learners. At the same time, it will be favorable to the improvement of website's topological structure and content update as well and reflects the dynamics and individualized more than ordinary distance education. At the research stage, the essay sufficiently combines theoretical research and practical application together.