PHGNet: Prototype-Guided Hypergraph Construction for Heterogeneous Spatiotemporal Forecasting
2026-05-25 • Artificial Intelligence
Artificial Intelligence
AI summaryⓘ
The authors focus on improving traffic forecasting, which helps manage city traffic by predicting future conditions. They note that existing methods struggle to understand complex interactions between many locations at once, so they create PHGNet. This new method groups similar traffic patterns together to better capture these relationships over time. Their approach includes ways to keep features consistent and refine predictions efficiently. Tests on real data show their method predicts traffic more accurately than previous ones.
traffic forecastingspatiotemporal dependencieshypergraphprototype learninghigh-order interactionsnode representationresidual refinementattention mechanismintelligent transportation systems
Authors
Ruiwen Gu, Yahao Liu, Zhenyu Liu, Qitai Tan, Xiao-Ping Zhang
Abstract
As a core task in intelligent transportation systems, traffic forecasting plays a critical role in urban traffic management. Accurate traffic forecasting relies on modeling complex spatiotemporal dependencies, which is inherently challenging due to spatial heterogeneity in traffic systems.Despite significant progress, most existing methods are still limited to pairwise spatial dependency modeling, making it difficult to capture dynamic high-order interactions among nodes with similar traffic patterns. To address this issue, we propose PHGNet, a novel spatiotemporal forecasting framework based on prototype-guided hypergraph construction. At the core of PHGNet, a prototype learning mechanism is designed to adaptively assign pattern-similar nodes to hyperedges, thereby capturing high-order interactions with time-varying structures. To improve the reliability of dynamic hypergraph construction, we further develop a global-local node representation module to extract time-consistent features. For forecasting, iterative residual refinement and Temporal Query Attention are introduced to improve forecasting accuracy while supporting efficient parallel decoding. Extensive experiments on multiple real-world datasets demonstrate that PHGNet achieves superior predictive performance compared with state-of-the-art methods.