GPS trajectory data segmentation based on probabilistic logic

With the rapid development of internet economy, transparent logistics is stepping into a prosperity period with massive transportation data generated and collected every day. In this paper, we focus on the segmentation of GPS trajectory data generated in logistics transportation to analyze the vehicle behaviors and extract business affair information according to the vehicle behavior characteristics, which is challenging due to the complexity of trajectory data and unavailability of road information. We extract the stopping points from the trajectory data sequence based on the duration of nonmovement, and construct business time window and electronic fence by analyzing the driving habits of vehicles. Furthermore, we propose a probabilistic logic based data segmentation method (PLDSM) which not only helps finding all the business points but also assists in inferring the business affair categories. An efficient numerical algorithm integrating duality theory and Newton's method is proposed to obtain the optimal solution. Finally, a practical example is presented to validate the effectiveness of PLDSM. The results greatly enrich the data segmentation technique and promote the practicability of probabilistic logic.