电信数据中用户行为特征测量与分析

Measurement and Analysis of User Behaviors in Mobile Data

通话和上网是电信运营商的重要业务,研究通话和上网的行为规律有助于提升电信运营商的业务规划和管理水平。现有的研究工作通常只关注于手机通话或上网行为,很少同时对两类行为进行关联的分析。该文提取了电信数据中手机通话与上网的基本特征,对通话和上网行为的频率分布进行了曲线拟合。通过比较两类行为的拟合参数与相关系数,发现了工作日与周末、以及周六与周日显著不同的用户行为特征。通过对通话和上网时间的归一化,定义了用户的使用偏好,发现54%的手机用户更多的倾向于使用手机通话,而31%的用户则倾向于使用手机上网。

Mobile calls and mobile-internet surfing are two important telecom services. The study of user behaviors on mobile calls and mobile-internet surfing is of great value to telecom operators in improving business planning and business management. Previous studies on user behaviors mainly focus on either mobile calls or mobile-internet surfing. There is little research that is conduced into the study of the interaction between mobile calls and mobile-internet surfing. In this paper, we first capture the basic features of the user behaviors in mobile calls and mobile-internet surfing. Through the curve fitting of frequency distributions of mobile calls and mobile-internet surfing, we find that there exist significant differences of user behaviors between workdays and weekends. We also normalize and compare the time that users spend on mobile-internet surfing and mobile calls. The results show that over 54% of users prefer using mobile calls, and over 31% of users prefer using mobile-internet surfing.