AIS Trajectory Partition Algorithm of Fishing Vessels Based on Complex Network and Community Detection

With the development of spatial data acquisition technology, trajectory data analysis based on location information has attracted widespread attention, which makes trajectory segmentation, as one of the basic steps of trajectory data mining, more critical. Proper segmentation of long-term trajectory data is conducive to the semantic enhancement of trajectory data, subsequent trajectory data calculation, and improvement of storage effectiveness. In this work, we focused on the partition of the AIS trajectory data of fishing vessels and proposed an AIS trajectory segmentation algorithm for fishing vessels based on complex network and community detection. In this algorithm, we first transformed the AIS trajectory data of the fishing vessels containing spatial-temporal information into a complex network through a complex network representation model, then used a community detection algorithm to find the community structure in the complex network and extracted the segmentation points, finally using the segmentation points to divide the trajectory data. We verified the proposed algorithm on real AIS trajectory data of fishing vessels and compared it with the trajectory segmentation benchmark algorithms CB-SMOT(Clustering-Based Stops and Moves of Trajectories) and GRASP-UTS (Greedy Randomized Adaptive Search Procedure for Unsupervised Trajectory Segmentation). The experimental results showed that the proposed algorithm achieves the optimum in the harmonic average of purity and coverage.