A Distributed Framework for Large-Scale Semantic Trajectory Similarity Join

With the development of 5G communications, location-based services (LBS) mobile applications have already become ubiquitous. As a result, applications like foursquare and Twitter have generated a massive amount of semantic trajectory data, which can no longer be efficiently processed by a single machine due to computing power limitations. Among the processing and analysis of semantic trajectory data, the similarity join query is a general yet computationally complex query operation. It is widely used in different applications, such as route planning, personalized carpooling, and geographic location recommendations. This paper proposes a distributed join framework to process similarity join over large sets of semantic trajectories. The framework supports the similarity join of semantic trajectories in the textual, temporal, and spatial domains. Experimental results on real datasets show that in comparison with baselines, the distributed framework in this paper has excellent scalability and query efficiency.