Online Shopping Preference Analysis of Campus Network Users Based on MapReduce

User online shopping preference mining is the key point on user found, e-commerce marketing and user personalized recommendation. A method for Online shopping preference analysis based on MapReduce is proposed in this paper. The campus network traffic is analyzed using MapReduce model, in which the features of user online shopping behavior are extracted by four MapReduce jobs using deep packet inspection (DPI). Making use of those features occuring to different e-commerce websites and with the help of the product information database established by a web crawler, user preference of e-commerce websites and categories of purchased product are analyzed. User conversion rates of three e-commerce websites(Taobao, Tmall, JD) are presented.