DWAFM: Dynamic Weighted Graph Structure Embedding Integrated with Attention and Frequency-Domain MLPs for Traffic Forecasting

2026-03-01

AI Summary

This work addresses the challenge of modeling dynamic spatiotemporal dependencies in traffic forecasting, a task often hindered by existing methods that either neglect graph structures or rely solely on static graphs. To overcome this limitation, the authors propose a novel Dynamic Weighted Graph Structure (DWGS) embedding approach, which introduces, for the first time, a dynamic graph capable of capturing time-varying association strengths between nodes. Building upon this, they develop a new model, DWAFM, that integrates spatiotemporal adaptive embeddings, time-feature embeddings, attention mechanisms, and a frequency-domain multilayer perceptron. Extensive experiments on five real-world traffic datasets demonstrate that DWAFM significantly outperforms state-of-the-art methods, achieving notably higher prediction accuracy.

Technology Category

• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems
• Machine Learning: Graph-based Machine Learning
• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning

Application Category

• Graph Algorithms and Modeling for the Web: Graph embeddings and representation learning for Web-related graphs
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

Accurate traffic prediction is a key task for intelligent transportation systems. The core difficulty lies in accurately modeling the complex spatial-temporal dependencies in traffic data. In recent years, improvements in network architecture have failed to bring significant performance enhancements, while embedding technology has shown great potential. However, existing embedding methods often ignore graph structure information or rely solely on static graph structures, making it difficult to effectively capture the dynamic associations between nodes that evolve over time. To address this issue, this letter proposes a novel dynamic weighted graph structure (DWGS) embedding method, which relies on a graph structure that can truly reflect the changes in the strength of dynamic associations between nodes over time. By first combining the DWGS embedding with the spatial-temporal adaptive embedding, as well as the temporal embedding and feature embedding, and then integrating attention and frequency-domain multi-layer perceptrons (MLPs), we design a novel traffic prediction model, termed the DWGS embedding integrated with attention and frequency-domain MLPs (DWAFM). Experiments on five real-world traffic datasets show that the DWAFM achieves better prediction performance than some state-of-the-arts.

Problem

• traffic forecasting
• spatial-temporal dependencies
• dynamic graph structure
• graph embedding
• node associations

Innovation

• Dynamic Weighted Graph Structure
• Graph Embedding
• Attention Mechanism
• Frequency-Domain MLPs
• Traffic Forecasting