Discovering Semantic Spatial and Spatio-Temporal Outliers from Moving Object Trajectories
This work addresses outlier detection in trajectory data for applications like urban planning or surveillance, but it is incremental as it builds on existing spatial outlier detection with semantic enhancements.
The paper tackled the problem of detecting outliers in moving object trajectories by introducing semantic spatial and spatio-temporal outliers, and proposed a new algorithm that finds such outliers not discovered by existing methods, as demonstrated with experiments on real data.
Several algorithms have been proposed for discovering patterns from trajectories of moving objects, but only a few have concentrated on outlier detection. Existing approaches, in general, discover spatial outliers, and do not provide any further analysis of the patterns. In this paper we introduce semantic spatial and spatio-temporal outliers and propose a new algorithm for trajectory outlier detection. Semantic outliers are computed between regions of interest, where objects have similar movement intention, and there exist standard paths which connect the regions. We show with experiments on real data that the method finds semantic outliers from trajectory data that are not discovered by similar approaches.