Enhancing short-term traffic prediction by integrating trends and fluctuations with attention mechanism

2025-04-28

AI Summary

Addressing the challenge of jointly modeling long-term trends and short-term fluctuations in short-term traffic flow forecasting, this paper proposes a dual-branch deep learning framework that separately captures trend (low-frequency) and fluctuation (high-frequency) components. A Bahdanau attention mechanism dynamically emphasizes critical time steps, enhancing sensitivity to transient events such as traffic congestion. Our key innovation lies in the first-ever trend–fluctuation dual-input parallel architecture, overcoming the detail loss inherent in conventional low-pass filtering approaches. Furthermore, we integrate a CNN–LSTM hybrid structure with multi-scale temporal feature decoupling and fusion to enable complementary representation learning. Experiments on 5–15-minute prediction horizons demonstrate significant improvements: R² increases by 12.7% and MAE decreases by 9.3% over state-of-the-art models. The framework effectively supports real-time congestion early warning and dynamic road network control.

Technology Category

• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems
• Machine Learning: Time-Series/Data Streams
• Search and Optimization: Learning to Search

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Web Mining and Content Analysis: Large pretrained models with web data

Abstract

Traffic flow prediction is a critical component of intelligent transportation systems, yet accurately forecasting traffic remains challenging due to the interaction between long-term trends and short-term fluctuations. Standard deep learning models often struggle with these challenges because their architectures inherently smooth over fine-grained fluctuations while focusing on general trends. This limitation arises from low-pass filtering effects, gate biases favoring stability, and memory update mechanisms that prioritize long-term information retention. To address these shortcomings, this study introduces a hybrid deep learning framework that integrates both long-term trend and short-term fluctuation information using two input features processed in parallel, designed to capture complementary aspects of traffic flow dynamics. Further, our approach leverages attention mechanisms, specifically Bahdanau attention, to selectively focus on critical time steps within traffic data, enhancing the model's ability to predict congestion and other transient phenomena. Experimental results demonstrate that features learned from both branches are complementary, significantly improving the goodness-of-fit statistics across multiple prediction horizons compared to a baseline model. Notably, the attention mechanism enhances short-term forecast accuracy by directly targeting immediate fluctuations, though challenges remain in fully integrating long-term trends. This framework can contribute to more effective congestion mitigation and urban mobility planning by advancing the robustness and precision of traffic prediction models.

Problem

• Improving traffic flow prediction by integrating long-term trends and short-term fluctuations
• Addressing deep learning limitations in capturing fine-grained traffic dynamics
• Enhancing short-term forecast accuracy using attention mechanisms for congestion prediction

Innovation

• Hybrid deep learning framework integrates trends and fluctuations
• Uses Bahdanau attention for critical time step focus
• Parallel processing of complementary traffic flow features