Channel attention-based spatial-temporal graph neural networks for traffic prediction
| Date | 03 May 2023 |
| Pages | 81-94 |
| DOI | https://doi.org/10.1108/DTA-09-2022-0378 |
| Published date | 03 May 2023 |
| Subject Matter | Library & information science,Librarianship/library management,Library technology,Information behaviour & retrieval,Metadata,Information & knowledge management,Information & communications technology,Internet |
| Author | Bin Wang,Fanghong Gao,Le Tong,Qian Zhang,Sulei Zhu |
Channel attention-based
spatial-temporal graph neural
networks for traffic prediction
Bin Wang , Fanghong Gao, Le Tong, Qian Zhang and Sulei Zhu
College of Information, Mechanical, and Electrical Engineering, Shanghai Normal
University, Shanghai, China
Abstract
Purpose –Trafficflow prediction has always been a top priority of intelligent transportation systems.
There are many mature methods for short-term trafficflow prediction. However, the existing methods are
often insufficient in capturing long-term spatial-temporal dependencies. To predict long-term dependencies
more accurately, in this paper, a new and more effective trafficflow prediction model is proposed.
Design/methodology/approach –This paper proposes a new and more effective trafficflow prediction
model, named channel attention-based spatial-temporal graph neural networks. A graph convolutional
network is used to extract local spatial-temporal correlations, a channel attention mechanism is used to
enhance the influence of nearby spatial-temporal dependencies on decision-making and a transformer
mechanism is used to capture long-term dependencies.
Findings –The proposed model is applied to two common highway datasets: METR-LA collected in Los
Angeles and PEMS-BAY collected in the California Bay Area. This model outperforms the other five in terms
of performance on three performance metrics a popular model.
Originality/value –(1) Based on the spatial-temporal synchronization graph convolution module, a spatial-
temporal channel attention module is designed to increase the influence of proximity dependence on decision-
making by enhancing or suppressing different channels. (2) To better capture long-term dependencies, the
transformer module is introduced.
Keywords Graph neural networks, Trafficflow forecasting, Spatial-temporal modeling, Attention
mechanism, Transformer, Time-series prediction
Paper type Research paper
1 Introduction
Temporal data contain huge application value, and intelligent analysis and prediction of
spatial-temporal data is an important issue for smart cities. However, due to the noise and
complex features of spatial-temporal data, the spatial-temporal prediction has always been
challenging.
In previous studies, various models often ignored non-Euclidean correlations determined
by complex road networks. To solve this issue, graph neural networks (GNN) and time-
series models are used to capture spatial and temporal correlations in transportation
networks, respectively.
Current spatial-temporal data prediction studies still cannot effectively capture long-
distance temporal dependencies. When the expansion rate of convolutional neural network
(CNN)–based methods increases, it is also easy to cause local signal loss spatial-temporal
graph convolutional network (STGCN; Yan et al., 2018). Recurrent neural network (RNN)–
based models can also be used for time-series prediction tasks (Elman, 1990), and their
variants long short-term memory (LSTM; Graves, 2012) and gated recurrent unit (Chung
et al., 2014) further enhance the ability of time-series modeling. However, the model based
on RNN/LSTM (Li et al., 2018;Wu et al., 2019) has always been unable to solve the time-
consuming problem and is prone to gradient disappearance or even gradient explosion
when capturing long-range dependencies.
ThecurrentissueandfulltextarchiveofthisjournalisavailableonEmeraldInsightat:
https://www.emerald.com/insight/2514-9288.htm
81
Received 28 September 2022
Revised 8 February 2023
31 March 2023
Accepted 8 April 2023
Data Technologies and
Applications
Vol. 58 No. 1, 2024
pp. 81-94
© Emerald Publishing Limited
2514-9288
DOI 10.1108/DTA-09-2022-0378
Graph neural
networks for
traffic
prediction
Get this document and AI-powered insights with a free trial of vLex and Vincent AI
Get Started for FreeStart Your Free Trial of vLex and Vincent AI, Your Precision-Engineered Legal Assistant
-
Access comprehensive legal content with no limitations across vLex's unparalleled global legal database
-
Build stronger arguments with verified citations and CERT citator that tracks case history and precedential strength
-
Transform your legal research from hours to minutes with Vincent AI's intelligent search and analysis capabilities
-
Elevate your practice by focusing your expertise where it matters most while Vincent handles the heavy lifting
Start Your Free Trial of vLex and Vincent AI, Your Precision-Engineered Legal Assistant
-
Access comprehensive legal content with no limitations across vLex's unparalleled global legal database
-
Build stronger arguments with verified citations and CERT citator that tracks case history and precedential strength
-
Transform your legal research from hours to minutes with Vincent AI's intelligent search and analysis capabilities
-
Elevate your practice by focusing your expertise where it matters most while Vincent handles the heavy lifting
Start Your Free Trial of vLex and Vincent AI, Your Precision-Engineered Legal Assistant
-
Access comprehensive legal content with no limitations across vLex's unparalleled global legal database
-
Build stronger arguments with verified citations and CERT citator that tracks case history and precedential strength
-
Transform your legal research from hours to minutes with Vincent AI's intelligent search and analysis capabilities
-
Elevate your practice by focusing your expertise where it matters most while Vincent handles the heavy lifting
Start Your Free Trial of vLex and Vincent AI, Your Precision-Engineered Legal Assistant
-
Access comprehensive legal content with no limitations across vLex's unparalleled global legal database
-
Build stronger arguments with verified citations and CERT citator that tracks case history and precedential strength
-
Transform your legal research from hours to minutes with Vincent AI's intelligent search and analysis capabilities
-
Elevate your practice by focusing your expertise where it matters most while Vincent handles the heavy lifting
Start Your Free Trial of vLex and Vincent AI, Your Precision-Engineered Legal Assistant
-
Access comprehensive legal content with no limitations across vLex's unparalleled global legal database
-
Build stronger arguments with verified citations and CERT citator that tracks case history and precedential strength
-
Transform your legal research from hours to minutes with Vincent AI's intelligent search and analysis capabilities
-
Elevate your practice by focusing your expertise where it matters most while Vincent handles the heavy lifting