In mobile commerce, user identification is a kind of necessary work for any marketing activity. However, as complexity of various factors, mobile users enter and exit communication network. When users re-enter the communication network, not only the records of old users are totally invalid, but we must re-start the work of inspecting new users and it will take a very long time to obtain the whole user profile again. So it is very valuable and important to do the user identification work.At present, we usually use the profile of user, IMEI code, the fingerprint of phone call to identify the user. However, the accuracy rate of the user identification work based on the user’s information and IMEI is not high, and it has many limitation based on the fingerprint of phone call, so we can not apply them to the real work.Therefore, this paper focuses on the problem of mobile user identification after they re-enter the communication network. And a similarity calculation method is presented based on users’ network access items and procedures. The method can identify a user by 7 steps from "pre-processing data, trimming the similar user set, identifying users".At last, to verify the feasibility and effectiveness of the presented method, we conduct an experiments on a real data set of the network access log of 25809 users in60 days. The overall accuracy rate of the presented identification method is 96.64%,... |