We Know What You Are--A User Classification Based on Mobile Data

With the rapid development of wireless communication technology and computer network technology, making calls through mobile handheld devices, i.e. mobile phone, sending SMS messages and getting access to wireless Internet have become the urgent needs of a large number of users. It seems to be quite necessary to understand and analyze the behavior pattern of wireless users. This paper aims to recognize the identities of users according to the rules and characteristics of mobile phone usage behaviors of users with different identities. This paper has collected the record of mobile phone usage behaviors from 140 users in Huazhong University of Science and Technology (including 70 teachers and 70 students), analyzed the behavior differences of these two types of users in making calls, sending SMS messages and surfing the Internet, and identified the users by machine learning algorithm according to the characteristics of user behaviors. The experimental results have verified the validity of user identity classification method proposed by this paper.