A model for semantic and spatio-temporal interconnected trajectory representation

In the era of big data, the full utilization of trajectory data has received extensive attention from researchers, who have proposed a large number of trajectory models with different characteristics and have been practical in terms of trajectory representation models. However, the existing trajectory representation models do not model the semantic features of trajectories sufficiently, and there is a lack of fusion between semantic features and spatio-temporal features, and there is no unified model expression. Therefore, there is an urgent need for a model that can represent the trajectory movement process to unify the semantic and spatio-temporal features of trajectory data. The model proposed in this paper constructs a trajectory domain ontology using knowledge graph technology under the premise of trajectory preprocessing and segmentation; then semantically enriches the trajectory entity by combining with open data; and finally verifies the practicability and validity of the model through the display of multi-dimensional features of the trajectory, the temporal deduction of the trajectory moving process, and the query of multi-dimensional features of the trajectory. This model does not need to develop complex data mining algorithms to realize the classification query and correlation analysis of multi-dimensional features of trajectory data.