The user behavior analysis is the effective way to enhance the user viscosity, maintain traffic flow among the large Internet network. This paper proposes the idea of user behavior analysis engine. It will combine the user behavior of static analysis with real-time acquisition of Web log. Using the data mining model based on cloud computing technology analyzes the contextual information that is obtained by accessing the page in Real-time, processing and storage at the same time. Experiments show that the system can improve the effect and efficiency of the user behavior analysis.