Font Size: a A A

Statistical Analisis And Modeling Of Calling Behavior

Posted on:2015-11-20Degree:MasterType:Thesis
Country:ChinaCandidate:Y H ZhouFull Text:PDF
GTID:2308330473453154Subject:Information security
Abstract/Summary:PDF Full Text Request
In recent years, due to the popularity of mobile devices, applying the mobile data in urban planning has attracted more and more attention. Mobile data includes not only the user’s call information, but also contains a user’s location information which can be used to analyze the user’s movement trajectory and call mode. Now because of the popularity of mobile phone, mobile data also implies the characteristics of urban areas. These characteristics can be found out to provide more comprehensive analysis of the city for urban planners and government policy makers, and they can make more rational urban planning. The thesis aims to analyze the call mode of user groups in areas covered by the base station and study characteristics of the city. The main work in the thesis is listed as follows:(1) Based on the communication network between the base stations, divide the city into different regions. Firstly, analyze the statistical characteristics of the network and find that the network has homogeneity for the administrative region. Secondly, use the Louvain algorithm to divide the network and find that the division is consistent with the actual administrative division.(2) Based on the communication network between the base stations, tracking the daily evolution of the city. Use the evolutionary algorithm based on Louvain algorithm to analyze the evolution of weekday and weekend day as well as urban communities at different times. Results show that whenever people tend to contact person near to them and the evolution of the community is related to the moving of people.(3) Based on the activity of the base station, identify the land uses in different parts of the city. Regard the activity of base station as a signal, adapt the correlation coefficient to measure the similarity of base station activities, and build a similarity network of base stations based on the similarity between the base station activities. By analyzing the community structure of the network, land uses of the base station coverage area were identified.The study about the communication network between base stations and the activity of base station appears to study the behavior of the base station, but actually to study the behavior of the user-groups in the areas covered by the base station. In real life, each base station corresponds to a coverage area and the use of this area is closely related to people’s activities. Focus on the base station, this paper analyzes the calling behavior from the perspective of groups and hope to understand the city dynamic and the use of urban facilities.
Keywords/Search Tags:call record, social network analysis, similarity, community detection, community evolution
PDF Full Text Request
Related items