Research On User Pattern Discovery In Web Usage Mining

Web usage mining (WUM) is one of the most important research fields in Web mining, relying on analysis and discovery of the rules in Web logs, it will find the potential customers of e-commerce and enhance the quality of Web service, moreover, it can improve the performance of the Web server. Compared with the current WUM methods, this paper has a discussion of the user pattern discovery based on Rough Set methodology. Cooperating with some existing algorithms, a user pattern discovery model will be established.In this paper, we analysis the characteristics of WUM, and the advantages and disadvantages of current methods, based on which we make a new way of data analyzing and modeling, and we improved and optimized the WUM process. We also discuss the parameters of the algorithms, and we put forward the different manipulation in different situations.We use a public WUM dataset to check the effectiveness of the system and make comparision with the former algorithms. As a result, our model is very effective and can make good prediction on user patterns.Finally, the problems requiring further studies are discussed.