Mining Mobile Internet Lifestyles in Distinct Urban Areas: Tales of Two Cities

Abstract

The information and communication technology industry has developed rapidly, and data are an important link between society and the Internet. The study of the regional usage features of mobile Internet users provides a guide for network supervision and service distribution. In this paper, we propose an approach to identify the typical mobile Internet patterns to account for the content of the user's Internet access, which is used to discover the online lifestyle of people living in cities of different sizes. We believe that the user Internet behavior is a weighted sum of many typical underlying features. Therefore, a single usage content tensor is constructed to record the sequence of content that the people access on the Internet. The mobile Internet lifestyles based on user context access content were obtained by comparing the composition of a user's typical online behavior under regional conditions. The method is applied to the case of two representative cities of different sizes in mainland China. We have extracted key features for the mobile Internet context for a large number of users in each city from the usage detail records obtained from cellular networks. This approach uncovers interesting features of human behavior on the mobile Internet at fine granularity, some of which allow us to quantitatively compare the online lives of people living in areas with different urban compositions. The identified regional typical user lifestyle can be used to guide urban planning, network supervision, service resource allocation, and urban dynamics analysis.

Keywords

• Lifestyles
• urban computing
• mobile Internet