Research On Web User Access Clustering Pattern

Along with Internet's rapid development, the enormous Web data resources have already become an important origin for people to obtain information. But, it is very hard for users to gain the true and valuable information rapidly and accurately due to the Web resources' traits, so the concept of data mining is presented. As a new technology, data Mining is applied to Decision Support System, and even can make prediction based on historic data. It can finally provide the convenient custom-made services to the users. Now it has already become a new and important research direction.This paper discusses the two phases, which are data preprocessing and clustering analysis, in Web transactions clustering analysis in a systematic and complete view. The data preprocessing phase also contains the procedures of log file interpretation, data washing, user identification and transaction identification; at the same time, in order to gain an easily interpreted result, the paper introduces the "Concept URL" in this phase. In clustering phase, a model of artificial ant is set up. Based on this model, the paper implements an ant colony clustering algorithm. What's more, k-means algorithm is also implemented in clustering analysis phase. The result is compared with that of ant colony algorithm. Experiment results are presented on web logs of a certain college to illustrate the techniques and methods. The quality of results is good.