The Topological Structure Of The Network That Web Excavates Is Upgraded And Studied

As a booming industry since 1990, Internet has already been a platform for human to work and study. WWW that accumulated billions of pages becomes a huge information repository. Thus, Internet is thought to be the 4-th industrial revolution of history. How to find valuable information from millions of web page rapidly and accurately and how to make website convenient for users to access turns into an important problem should be settled imminently.Data Mining is the process that people discover connotative, unknown and valuable information or mode from large database or data warehouse, which combines theory and technology of multi-fields, such as database, artificial intelligence, machine learning, statistics and so on. Web Mining is one of hotspot of Data Mining technology applied in artificial intelligence field. It is a more defiant subject that implements some function as discovering web access mode, web structure and rule and dynamic search for web content. Web usage mining can find useful user access pattern and corresponding potential customers from large volume of use access logs, so that corporations would provide personal service and make their electronic commerce strategies.The paper mainly researches applying web usage mining in improving website topological structure.Firstly, it describes the Data Mining techniques concepts, methods, techniques and in general, interprets what is Data Mining, why Data Mining, how to Data Mining and introduces its development and application. Based on analysis of the Web Log data and our research of the difficulty and problems of current method, we bring forward PCWS, a model composed of Preprocessing, Classifier, Watcher and Synthesizer to improve website topological structure. It makes advantage of present web mining technology. Then we introduce PCWS in details and use this model to carry on the log pretreatment, users discern, conversation discerns, find user's route and recognizing sequence-pattern. Finally the result of the experiment is evaluate.With success experimentally, the system has reached its goal of experiment and study, which establish the foundations for the further research of web mining.