Research And Implementation Of User Behavior Analysis System For Mobile E-commerce

The emergence and rapid development of e-commerce has greatly facilitated the living of society, while it brings new challenges and opportunities to the mobile e-commerce enterprises and the operators. On the one hand, the enterprises are facing serious homogeneous competition, which leads to the most important issue for mobile e-commerce enterprises about how to efficiently grasp the market demand. On the other hand, in order to seize the opportunities provided by mobile e-commerce, the operators must make full use of existing network resources and user data to provide new business and personalized business. Therefore, it is significant for mobile e-commerce enterprises and operators to deeply dig up the requirements of users by researching and analyzing of user behavior for mobile e-commerce.Focused on e-commerce behavior of mobile Internet users, a set of user behavior system for mobile e-commerce is researched and implemented in this thesis. The system achieves the precise identification and statistical analysis on the mobile e-commerce user behavior data, and obtains user behavior characteristics and interest preferences. Three main aspects are included in this thesis as follows:1. Based on the key technologies of business identification and requirements of the system, an overall design scheme of user behavior analysis system for mobile e-commerce is proposed. The system includes four modules, namely acquisition/processing, fine identification of business, statistical analysis of business data and management of the application. Moreover, the fine identification of business and statistical analysis of business data are discussed significantly in this thesis.2. Combined with the overall architecture of the scheme, the two main modules of the system are proposed. In the fine identification of business, feature library is established through packet snatch and web crawler firstly, then the first DPI processing and the second DPI processing are employed to achieve the identification of the type and content of business, and thus to obtain the user s’ complete information. In statistical analysis of business, data processing is analyzed firstly, then the related table structure in the system is designed, and data loading program is achieved. Furthermore, by programming storage procedure, user behavior analysis, including browsing content analysis, hot commodity analysis, terminal analysis and preference analysis are achieved.3. The function of the system is verified by collecting the real-time measured data of the present network, and the results of user behavior analysis are displayed on the client, which shows the specific behavior characteristics and the preference of the users. At present, the system has been successfully applied to a big data analysis platform in mobile Internet by a certain communication technology company. Meanwhile, the system runs steadily and efficiently, and has achieved the expected goal.