CLGSDN: Contrastive Learning Based Graph Structure Denoising Network for Traffic Prediction

The Graph Neural Network-based prediction models have demonstrated remarkable utility in traffic prediction, and their efficacy is highly determined by the quality of the provided graphs. Consequently, there is an increasing demand for employing Graph Structure Learning (GSL) techniques to optimize...

Full description

Saved in:
Bibliographic Details
Published in:IEEE internet of things journal 2024-11, p.1-1
Main Authors: Peng, Peng, Chen, Xuewen, Zhang, Xudong, Tang, Haina, Shen, Hanji, Li, Jun
Format: Article
Language:English
Subjects:
Adaptation models
Contrastive learning
Graph Contrastive Learning
Graph Generation
Kernel
Noise
Noise measurement
Noise reduction
Optimization
Predictive models
Probabilistic logic
Probability Modeling
Spatiotemporal phenomena
Traffic Prediction
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:The Graph Neural Network-based prediction models have demonstrated remarkable utility in traffic prediction, and their efficacy is highly determined by the quality of the provided graphs. Consequently, there is an increasing demand for employing Graph Structure Learning (GSL) techniques to optimize or generate the graphs. However, existing GSL techniques for traffic prediction encounter various issues, including the absence of temporal dynamicity, noisy connections, and insufficient supervisory information. To address these limitations, this paper proposes a novel two-stage graph generation framework called Contrastive Learning-based Graph Structure Denoising Network (CLGSDN). This framework formulates the graph generation task as a probabilistic observation-inference process: using self-learning adjacency matrix and Time Delayed Self-Attention (TDSA) methods to generate a series of graph observations, then inferring the optimal graph based on observations. The self-learning adjacency matrix is responsible for learning all potential connections in the graph, while TDSA enables the graph to change with traffic flow. In addition, CLGSDN identifies and eliminates noisy connections by modeling negative samples of the graph (edges), and defines virtual labels to achieve Spatiotemporal Graph Contrastive Learning (ST-GCL) in traffic prediction. The experimental results show that CLGSDN significantly enhances current mainstream traffic prediction models by providing reliable and efficient graphs. As such, it has significant implications for a wide range of applications, including traffic management, logistics, and smart transportation systems.
ISSN:2327-4662
2327-4662
DOI:10.1109/JIOT.2024.3502517