Similarity Measurement of Trajectory Data Stream Based on Incremental DBSCAN

With the rapid development of mobile internet and devices, a large amount of trajectory data of moving objects has been generated. The similarity measurement of trajectory data plays a crucial role in exploring valuable information, such as human mobile behavior and patterns. However, most existing work constructs similarity measurement offline, which cannot provide timely feedback for time sensitive applications. In light of this, we present a novel hierarchical framework based on incremental DBSCAN in this paper to measure the similarity between users in real-time. Our proposal includes a hierarchical directed graph framework (HDGF) to ensure the accuracy of trajectory similarity measurement and a stream processing algorithm based on incremental DBSCAN (IDBSCAN) to update the hierarchical framework of users in real-time. Experiment results on real trajectory dataset show that our proposed method is effective and offers competitive performance compared with related methods.