Analysis Of Online And Offline Behavior Based On Mobile Network Mass Traffic

With the development of the mobile Internet,the mobile Internet traffic data has also increased to an unprecedented scale.The era of big data is coming.Traditional telecommunications operators are facing new opportunities and challenges in the era of big data.On the one hand,the demographic dividend of the telecommunications market has come to an end.On the other hand,these operators are stocked with a large number of customer related data.How to excavate more valuable information in these data,and improve user experience is an urgent task for operators.It is also an important historical task related to the national economy and the people's lives.The study of human mobility which aims at revealing the general laws of human movement is the basis of many social,economic,and technological phenomena.It has received wide attention.The research on the online browsing mode of mobile Internet users not only helps to better understand human behavior,but also has high commercial value.At present,the research on the relationship between user behavior and mobile browsing behavior is very limited,and only stops in finding the relationship between certain types of mobile applications and geographic location or mobile pattern.This thesis tries to explore the relationship between the mobile pattern and the online browsing pattern from the perspective of the user's living habits.To achieve this goal,we need to deal with several challenges,including the effect of data size on the credibility of experimental results,the identification of urban function areas,the modeling of user mobility,and the modeling of user browsing tendency.This thesis uses a 7 day real mobile Internet data set in a northern province of China,covering 181873 users' online behavior.In the field of experimental tools,we use the self-developed data analysis platform to analyze the data.In terms of experimental methods,we have absorbed some advantages of the existing recognition methods of urban functional areas,and improved the robustness of this algorithms from multiple perspectives.In addition,we propose a novel spatial-temporal mobility modeling method--user mobile image,which has the advantages on computation and display.Finally,we analyze the relationship between the user online and offline behavior.Experimental results show that the proposed method of urban function area recognition performs well on real mobile Internet data sets,and paved the way for the identification of urban functional areas by large-scale mobile Internet data sets.In addition,the user mobile image modeling method proposed in this thesis can carry more information compared with the location based mobility model,which is more meaningful for predicting online browsing mode.