基于核心网数据挖掘的移动通信用户行为分析及应用

发布时间：2018-05-25 20:28

本文选题：核心网 + 数据挖掘　；参考：《北京邮电大学》2014年硕士论文

【摘要】：随着移动通信技术的飞速发展,移动通信用户规模不断扩大,运营商的网络建设也趋于成熟。在网络运行维护过程中,运营商积累了海量的数据,并且以指数级规模不断增长。这些数据,尤其是核心网的数据中,蕴含着丰富的用户行为信息。因此如何从这些大数据中快速并精准地挖掘出对运营商有价值的用户行为信息,是运营商普遍面临的一个重要课题。在竞争日趋激烈的通信运营市场中,准确掌握用户行为信息对于运营商来说有极其重要的意义。本论文针对于核心网的多种数据源,提出了一套普适的数据挖掘流程,结合移动通信用户行为的特殊性,尝试提出了一般的用户行为分析方法,然后结合运营商的现网采集数据进行了实际的数据挖掘操作,并且对挖掘结果进行了一系列的业务场景分析。本论文的研究工作和成果如下： 1.基于核心网多种常见数据源的特点,建立了专门针对核心网数据源的数据挖掘流程。 2.通过分析移动通信用户的一般行为特点,提出了用户行为分析的基本理论和方法。 3.基于现网的实际数据,运用聚类等一系列数据挖掘方法进行数据挖掘的实现工作,得到了一系列有价值的用户行为信息。 4.将己知的用户行为在一定适用范围内进行数值建模,并且深入研究了数值模型的实际含义。
[Abstract]:With the rapid development of mobile communication technology, the scale of mobile communication users continues to expand, and the network construction of operators tends to mature. In the process of network operation and maintenance, operators accumulate massive data and increase exponentially. These data, especially in the core network, contain rich user behavior information. Therefore, how to quickly and accurately mine the valuable user behavior information from these big data is an important topic that operators generally face. In the increasingly competitive communication operation market, it is very important for operators to accurately grasp user behavior information. In this paper, a set of universal data mining flow is proposed for multiple data sources in the core network. Combined with the particularity of mobile communication user behavior, a general user behavior analysis method is proposed. Then the actual data mining operation is carried out based on the data collected by the operator, and a series of business scenarios are analyzed. The research work and results of this thesis are as follows: 1. Based on the characteristics of many common data sources in the core network, a data mining process for the core network data sources is established. 2. By analyzing the general behavior characteristics of mobile communication users, the basic theory and method of user behavior analysis are put forward. 3. Based on the actual data of the present network, a series of data mining methods, such as clustering, are used to realize the data mining, and a series of valuable user behavior information are obtained. 4. The known user behavior is numerically modeled within a certain range of applications, and the actual meaning of the numerical model is studied in depth.
【学位授予单位】：北京邮电大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TN929.5;TP311.13

【参考文献】