The Prediction of User State for Mobile Device

摘要：

The DM management system has many operations to device software, and the schedule is an important aspect in the process of operations. Because many management operations need to determine the implement time based on the users states, learning the variation of users states in advance is of importance to make a reasonable schedule. In this article, the user states comprise online, busy, away, invisible and offline on the basis of mobile device service firstly. And then we import the continuous time Markov chain, and each transition probability between two states at some time can be computed through the algorithm, then we can gain the probability of some state at this time, ultimately, the DM operations can be scheduled reasonably in advance according to the value of the probability. In this article we also have a simulation based on Markov algorithm for the prediction model and confirm it in our Device Management System.

关键词： Prediction User State Mobile Device

作者: LiJuan Men ShaoYong Guo LanLan Rui XueSong Qiu

作者单位: State Key Laboratory of Networking and Switching Technology Beijing University of Posts and Telecommunications Beijing 100876 China

会议类型: 国际会议

会议名称: 2010 International Conference on Future Information Technology(2010年未来信息技术国际会议 ICFIT 2010)

会议地点: 长沙

会议语种:英文

页码: 348-353

在线出版日期: 2010-12-14（万方平台首次上网日期，不代表论文的发表时间）