Research on Accompanying Vehicle Recognition Based on Big Data Technology

With the acceleration of urbanization, the transportation system is facing increasingly complex governance challenges, among which the identification and analysis of accompanying vehicle behavior has become a key issue. Traditional methods have some defects, such as low efficiency and limited coverage, and it is difficult to deal with massive data. In this study, an accompanying vehicle identification framework based on big data technology is constructed. By efficiently processing massive trajectory data at city level, fusing multi-source heterogeneous information and designing extensible algorithms, real-time accurate identification of accompanying behavior is realized. A hybrid model combining spatio-temporal feature engineering and graph neural network (GNN) is proposed to effectively describe the spatio-temporal similarity between vehicles and capture the complex dynamic adjoint relationship. The experimental results show that the framework has excellent performance in the accompanying vehicle identification task, which can provide technical support for public safety applications such as deck car tracking and gang crime warning, and also provide theoretical and practical reference for intelligent transportation system optimization and spatio-temporal data mining.