Traffic Flow Prediction Model Based on Variant Hybrid Multi-Hop Graph Convolution

Accurate traffic flow prediction is a crucial step in building an intelligent transportation system, and it is of great significance for alleviating urban traffic congestion and optimizing travel routes. Due to the complex spatial topology of the transportation network and the highly nonlinear temporal dynamic characteristics of the flow data, traditional prediction methods are difficult to fully capture the inherent spatio-temporal dependencies. Therefore, this paper proposes a traffic flow prediction model based on variant hybrid multi-hop graph convolution. Firstly, by introducing a multi-hop graph convolution operator, the model explicitly aggregates the spatial information of multiple-order neighborhoods in the traffic sensor network to capture long-distance spatial dependencies. Secondly, a variant hybrid graph convolution module is designed, which combines Chebyshev polynomial approximation and adaptive adjacency matrix, while balancing computational efficiency and enhancing the model's ability to mine implicit spatial correlations. In the temporal dimension, we adopt a method combining gated temporal convolution and attention mechanisms to dynamically capture the temporal change patterns of traffic flow. Evaluations across public datasets reveal that our method demonstrates consistent advantages when benchmarked against baseline approaches. The model achieves better performance in short-term and medium-term traffic flow prediction tasks, significantly reducing prediction errors and verifying the effectiveness and advancement of the proposed model.