Mining Daily Behavior Patterns from Longitudinal Context Data on Smartphones
Various sensors embedded in smartphones can collect a large quantity of raw data every day.With these logged data, multiple context covering most aspects of a users daily life ranging from locations to app usage can be inferred.We design a framework to mine users daily behavior patterns on smartphones.To aggregate multi-modal context data inferred from raw sensor data, we propose a dynamic sliding window approach.Moreover, we develop a frequent patterns mining algorithm which takes both frequency and duration of context occurrence into consideration.By conducting experiments on 21 users over 6 weeks, it is proved that our framework is feasible and efficient when running on resources limited smartphones.Whats more, we find out a lot of useful and interesting daily behavior patterns which reflect users lifestyle.At last, we visualize the patterns from two perspectives: behavior patterns in different locations and time periods.
Mobile Data Mining Smartphone Sensing Behavior Pattern
Han Li Dianxi Shi Ruosong Yang Xiaoyun Mo
National Laboratory for Parallel and Distributed Processing, School of Computer,National University of Defense Technology, Changsha, Hunan 410073
国际会议
三亚
英文
430-435
2015-12-26（万方平台首次上网日期，不代表论文的发表时间）