通话规律的统计分析和建模

发布时间：2018-07-08 21:05

本文选题：手机数据 + 社会网络　；参考：《电子科技大学》2014年硕士论文

【摘要】：近年来,随着手机设备的逐渐普及,手机数据在城市规划中的应用受到了越来越多的关注。手机数据不仅包含用户的通话信息,而且还包含用户的位置信息,可以用于分析用户的移动轨迹和通话模式。由于手机设备的普及,手机数据中几乎包含了所有人的通话和位置信息,现在手机数据还可以用于分析城市的行为特征。这些行为特征可以被挖掘出来,给城市规划者和政府决策者提供更为全面的城市分析结果,从而做出更合理的城市规划。本文旨在分析基站覆盖区域内用户群体的通话模式,研究城市的行为特征。本文的工作概括为以下几点:(1)基于基站间的通话联系图,对城市进行了区域划分。首先分析了基站间通话图的网络统计特征,发现该网络具有行政区域性质的同质性。其次,采用Louvain算法对该网络进行了划分,划分结果在地图上的分布与实际的行政区域划分比较一致。(2)基于基站间的通话联系图,对城市的社区演变过程进行了跟踪。采用一个基于Louvain算法的社团演化跟踪算法,分析了工作日和周末、每天不同时刻城市的社区演变过程。实验结果表明,无论何时人们都倾向于联系距离自己近的人,并且发现社区的演化与人们的移动有关。(3)基于基站的活动,对城市不同地区的土地利用情况进行了识别。将基站的活动看作一个信号,用相关系数来度量基站活动的相似度,并根据基站活动的相似度构建了基站间的相似关系网。通过分析该相似关系网的社团结构,对基站覆盖区域的土地利用类型进行了识别。基站间的通话图以及基站的活动表面上似乎是研究基站的行为,其实研究的是基站覆盖区域的用户群体的行为。实际生活中,每一个基站都对应着一个覆盖区域,该区域的使用情况与该区域下人们的活动息息相关,因此本文以基站为研究对象,从群体的角度分析人们的通话行为,希望可以了解到城市的变化和城市设施的使用情况。
[Abstract]:In recent years, with the popularity of mobile devices, the application of mobile phone data in urban planning has received more and more attention. The mobile phone data not only contains the user's call information, but also contains the user's location information, which can be used to analyze the user's mobile track and call mode. Because of the popularity of mobile devices, mobile phone data almost contain all the phone and location information, now mobile phone data can also be used to analyze the behavior characteristics of the city. These behavioral characteristics can be excavated to provide urban planners and government decision makers with more comprehensive urban analysis results and thus to make more rational urban planning. The purpose of this paper is to analyze the call modes of user groups in the base station coverage area and to study the behavior characteristics of the city. The work of this paper is summarized as follows: (1) based on the telephone contact diagram between the base stations, the area of the city is divided. Firstly, the network statistical characteristics of the call graph between base stations are analyzed, and it is found that the network has the homogeneity of administrative region. Secondly, the network is partitioned by Louvain algorithm, and the distribution on the map is consistent with the actual administrative division. (2) based on the telephone contact diagram between base stations, the evolution process of the city community is tracked. A community evolution tracking algorithm based on Louvain algorithm is used to analyze the process of community evolution in cities at different times of working day and weekend. The experimental results show that people tend to contact people who are close to them at any time, and find that the evolution of communities is related to people's movement. (3) based on the activities of base stations, land use in different areas of cities is identified. The activity of base station is regarded as a signal, and the similarity of base station activity is measured by correlation coefficient, and the similarity relation network between base stations is constructed according to the similarity of base station activity. By analyzing the community structure of the similar network, the land use types of the base station coverage area are identified. On the surface, the call diagram and the activity of the base station seem to study the behavior of the base station, in fact, what is studied is the behavior of the user group in the area 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 the activities of people in the region. Therefore, this paper takes the base station as the research object and analyzes the communication behavior of people from the point of view of the group. Hope to know about the changes in the city and the use of urban facilities.
【学位授予单位】：电子科技大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TN929.5

【相似文献】