Data Mining Research On The User' Web Log Behavior

Along with the continuously mature development of data mining technology, the use of data mining technology to mining was studied for the campus network which has great application value, the user's log analysis of Internet behavior is the main direction of network research and management. So this article mainly studies the content of the online behavior of the number of students which is to analyze the campus network, digging out the user's online time, type of user access to sites such as the Internet behavior characteristics and so on. Aiming at controlling and predicting the running status of campus network by analyzing the behavior of students and the regularities of the logging which can gives the basis of scientific management to the construction of campus networkThe detailed work is shown as following:Firstly, there is a simple overview for the working of systematic analysis, for campus network users in the online journal data collection, is obtained by data preprocessing, data transformation is suitable for the analysis of the data attribute set, the data stores in the database which is used for mining.Then this article does detailed introduction on the association rules Apriori algorithm, and analyzes the deficiency of the algorithm. Referring to the negative association rules, combine the two algorithms together, analyzes its feasibility, and put forward own improvement ideas, to improve the running speed of the algorithm, and the correctness of the improved algorithm is verified by experimental contrast.Finally campus students surf the Internet behavior is to be studied from many aspects, including namely the minute online number analysis, user access to sites, different categories of users surfing the Internet behavior analysis and so on. Among them, the analysis of the user to access the website and the analysis of different categories of users surfing the Internet, use Weka data mining the association rules algorithm in the platform were analyzed, and the Internet characteristics of the customers can be obtained. By analyzing students' Internet behavior, reasonable suggestions on optimization of campus network is given. This research can be applied in the field of business, to provide customers with personalized service, enhance the competitiveness of the enterprises.