NSIM: A robust method to discover similar trajectories on cellular network location data
Yupeng Tuo, Xiaochun Yun, Yongzheng Zhang · 2017
Trajectory analysis is crucial and has been more and more widely used in various fields, such as location-based services, urban traffic control, route plan, etc. The existing methods have certain limitations when applied to cellular network location data. In this paper, we propose NSIM, a novel approach that can effectively discover similar trajectories. In NSIM, we first design an algorithm that can discover all the common moving patterns among trajectories, and then we adopt a vectorization method to abstract each trajectory as a summary vector that composed of specific common moving patterns. Finally we measure the similarity of trajectories by computing the distance between the summary vectors. Extensive experiments on real-world dataset show that, compared with three other approaches, NSIM achieves good effectiveness in most test cases and achieves better efficiency when applied to small or medium length trajectories.