Research On Algorithm Of Web User Browsing Pattern Fuzzy Clustering

Web using mining tries to discover interesting web user access patterns or knowledge from the web log records. Analyzing and exploring regularities in web log records can identify potential customers for electronic commerce, enhance the quality and delivery of Internet information services to the end web user, and improve web server system performance. Web mining, especially web users clustering of web usage mining, has obvious fuzzy characteristic, so fuzzy clustering is sometimes better suit for the web mining in comparison with traditional clustering. A mothed based on maximum tree algorithm for fuzzy clusering is brought out in this thesis to cluster the web session.First, a survey of data mining and web data mining technique is given. The content of study and development of web mining is emphasized.Second, this thesis analyses the disadvantages of algorithm in existence. The idea that applies fruzzy clustering to web mining is affirmed. Afterward, the describing material algorithm based maximum tree algorithm for fuzzy clusering is given. And an instance of algorithm is brought forward to expound this algorithm.Finally, correctness of this algorithm is proved in theory. And this algorithm is proved to have better accurary, fewer CPU time and better acalability than others by the analysis of performance. It can be used in the field of E-business, such as personalized Web and Web recommendation.