Research And Implementation Of Mining Implicit User Interest

As a method of knowledge discovery, data mining has been widely used, and was the most active domain of database. Web mining is to use the traditional data mining technologies to extract information and knowledge in the Web environment. The web usage mining is the most widely used method, which is used in the field of e-commerce, internet ads, intelligent recommendation system, internet marketing, and intelligent decision support. A good model of web mining is the key to the success of web usage mining. This dissertation will focuse on the research of implicit user interests mining. By using the data mining technologies on the documents which one user has visited, we can establish a user interest model for the user. Furthermore, the user model can be used to provide some personalized services.In this paper, we will first concern the development and main technologies of web usage mining, as one part of web usage mining; we will especially concern the establishment of user interest model. By comparing of the use of text categorization and text clustering technologies in user interest mining, we present a user interest model based on text clustering.Data preprocessing is the preparation stage of web mining. In this dissertation, we will introduce some key technologies, including the filtering of web log and web page content extraction, and then we propose a new system integration method based on the technology of pipeline.Research of the text clustering technology which is applicable to user interest mining is the core content in this paper. We will first survey the main text clustering methods and demonstrate their respective characteristics. By analysis of the requirements of user interest mining, we choose the BIRCH method to cluster one user's visited documents to establish the interest model.In the last, we demonstrate a user interest mining system based on the web log from Myspace China, and we do some experiments to the text clustering method we used.