Contrastive Trajectory Similarity Learning with Dual-Feature Attention
This work addresses trajectory similarity queries in databases, offering a more accurate and efficient solution for applications like location-based services, but it is incremental as it builds on existing contrastive learning and attention mechanisms.
The paper tackled the problem of efficiently and accurately measuring trajectory similarity for database queries by proposing TrajCL, a contrastive learning-based method with dual-feature self-attention, which significantly outperformed state-of-the-art measures and achieved up to 56% higher accuracy after fine-tuning.
Trajectory similarity measures act as query predicates in trajectory databases, making them the key player in determining the query results. They also have a heavy impact on the query efficiency. An ideal measure should have the capability to accurately evaluate the similarity between any two trajectories in a very short amount of time. Towards this aim, we propose a contrastive learning-based trajectory modeling method named TrajCL. We present four trajectory augmentation methods and a novel dual-feature self-attention-based trajectory backbone encoder. The resultant model can jointly learn both the spatial and the structural patterns of trajectories. Our model does not involve any recurrent structures and thus has a high efficiency. Besides, our pre-trained backbone encoder can be fine-tuned towards other computationally expensive measures with minimal supervision data. Experimental results show that TrajCL is consistently and significantly more accurate than the state-of-the-art trajectory similarity measures. After fine-tuning, i.e., to serve as an estimator for heuristic measures, TrajCL can even outperform the state-of-the-art supervised method by up to 56% in the accuracy for processing trajectory similarity queries.