Design And Inplementation Of Mobile Internet User Behavior Analysis System Based On Telecom Big Data

In the mobile Internet Era, the development of techno logy brings great convenience for people. Telecommunications operators as an important channel provider of mobile Internet, master the pipeline circulation of massive data, which contains a very commercial value of user access behavior information. The key problem for telecommunications operators to improve their service value is how to study user's Internet access behavior and traffic usage through big data processing technology,discover user surfing habits and preferences, and insight into user needs.Although there are many researches on the analysis of Internet users'behavior, but most of the current user behavior analysis platform provides statistical analysis services or provide fast query by building the data warehouse. A large data mining platform for automated analysis of telecom user behavior is needed.Under the background, this paper designed a mobile Internet user behavior analysis system for telecom big data based on the requirements of mobile Internet user behavior analysis. Around the related technology,system requirements, system design and implementation, scene analysis and system testing, the main contents of this paper include: Firstly, based on the requirements of mobile users' Internet data processing and the methods of mobile Internet user behavior analysis, a mobile Internet user behavior analysis system is designed which has data management, data preprocessing, data statistical analysis, data mining calculation and data visualization and other functions. Secondly, the bottom of the system uses a distributed cluster composed of HDFS and Spark, providing the storage and processing capabilities of big telecom data. And web page operations and data visualization display provides simple and clear data processing and analysis methods for users. Thirdly, the system provides statistical analysis and data mining methods, and user behavior analysis and user flow clustering analysis of two scene analysis, realize user traffic clustering based on the K-means algorithm. The research on user time distribution,service and application preferences, traffic characteristics satisfies the basic needs of mobile Internet user behavior analysis and mining.The mobile Internet user behavior analysis system provides mobile Internet users' data processing, analysis operations and data visualization display. An accurate and efficient platform is provided for telecommunications operators to achieve insight into user needs, precise user positioning and meticulous operation.