关键词: 物联网, 用户特征, 时空特性, GPS轨迹, 序列挖掘

Abstract: Internet of Things(IoT)encompasses a huge number of users and physical entities,which usually have temporal and spatial features.User spatiotemporal profile is the key prerequisites to build personalized applications in IoT.Therefor,a user spatiotemporal profile mining algorithm was proposed to discover travel patterns from user Global Positioning System(GPS)trajectories.A GPS trajectory was transformed into a sequence of Regions of Interest(ROI)based on spatial and temporal property of GPS points.Then a pattern-growth mining algorithm was proposed to mine asynchronous periodic sequential patterns with multiple minimum item supports,which were not only occurring frequently,but also appearing periodically.The experimental results showed the efficiency and accuracy of the proposed algorithm.

Key words: Internet of things, user profile, spatiotemporal characteristics, global positioning system trajectories, sequential pattern mining