Short-term urban traffic forecasting in smart cities: a dynamic diffusion spatial-temporal graph convolutional network

Abstract Short-term traffic forecasting is an important part of intelligent transportation systems. Accurately predicting short-term traffic trends can avoid traffic congestion and plan travel routes, which is of great significance to urban management and traffic scheduling. The difficulty of short-...

Full description

Saved in:
Bibliographic Details
Main Authors: Xiang Yin, Junyang Yu, Xiaoyu Duan, Lei Chen, Xiaoli Liang
Format: Article
Language:English
Published: Springer 2025-01-01
Series:Complex & Intelligent Systems
Subjects:
Online Access:https://doi.org/10.1007/s40747-024-01769-6
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1823861510717308928
author Xiang Yin
Junyang Yu
Xiaoyu Duan
Lei Chen
Xiaoli Liang
author_facet Xiang Yin
Junyang Yu
Xiaoyu Duan
Lei Chen
Xiaoli Liang
author_sort Xiang Yin
collection DOAJ
description Abstract Short-term traffic forecasting is an important part of intelligent transportation systems. Accurately predicting short-term traffic trends can avoid traffic congestion and plan travel routes, which is of great significance to urban management and traffic scheduling. The difficulty of short-term urban traffic forecasting is that the traffic flow is random and will be dynamically changed by the traffic conditions of nearby nodes. In order to solve this problem, this paper proposes a model based on Dynamic Diffusion Spatial-Temporal Graph Convolutional Network. It first combines the dynamic generation matrix and the static distance matrix to grasp real-time traffic conditions, and then introduces the diffusion random walk strategy to capture the correlation of spatial nodes. Finally, the convolutional LSTM module is used to mine the spatiotemporal dependence of traffic data to improve the accuracy of traffic prediction. Compared to several baseline models, the experimental results show that the model is 7% better than other models on several metrics and demonstrates the necessity of the module through ablation experiments.
format Article
id doaj-art-df6d1d1cec884f2da03ef260de6bd954
institution Kabale University
issn 2199-4536
2198-6053
language English
publishDate 2025-01-01
publisher Springer
record_format Article
series Complex & Intelligent Systems
spelling doaj-art-df6d1d1cec884f2da03ef260de6bd9542025-02-09T13:01:23ZengSpringerComplex & Intelligent Systems2199-45362198-60532025-01-0111211210.1007/s40747-024-01769-6Short-term urban traffic forecasting in smart cities: a dynamic diffusion spatial-temporal graph convolutional networkXiang Yin0Junyang Yu1Xiaoyu Duan2Lei Chen3Xiaoli Liang4School of Software, Intelligent Data Processing Engineering Research Center of Henan Province, Henan UniversitySchool of Software, Henan UniversitySchool of Computer Science, Wuhan UniversitySchool of Software, Henan UniversitySchool of Software, Henan UniversityAbstract Short-term traffic forecasting is an important part of intelligent transportation systems. Accurately predicting short-term traffic trends can avoid traffic congestion and plan travel routes, which is of great significance to urban management and traffic scheduling. The difficulty of short-term urban traffic forecasting is that the traffic flow is random and will be dynamically changed by the traffic conditions of nearby nodes. In order to solve this problem, this paper proposes a model based on Dynamic Diffusion Spatial-Temporal Graph Convolutional Network. It first combines the dynamic generation matrix and the static distance matrix to grasp real-time traffic conditions, and then introduces the diffusion random walk strategy to capture the correlation of spatial nodes. Finally, the convolutional LSTM module is used to mine the spatiotemporal dependence of traffic data to improve the accuracy of traffic prediction. Compared to several baseline models, the experimental results show that the model is 7% better than other models on several metrics and demonstrates the necessity of the module through ablation experiments.https://doi.org/10.1007/s40747-024-01769-6Traffic forecastingGraph convolutional networkSpatial-temporalDynamic generation
spellingShingle Xiang Yin
Junyang Yu
Xiaoyu Duan
Lei Chen
Xiaoli Liang
Short-term urban traffic forecasting in smart cities: a dynamic diffusion spatial-temporal graph convolutional network
Complex & Intelligent Systems
Traffic forecasting
Graph convolutional network
Spatial-temporal
Dynamic generation
title Short-term urban traffic forecasting in smart cities: a dynamic diffusion spatial-temporal graph convolutional network
title_full Short-term urban traffic forecasting in smart cities: a dynamic diffusion spatial-temporal graph convolutional network
title_fullStr Short-term urban traffic forecasting in smart cities: a dynamic diffusion spatial-temporal graph convolutional network
title_full_unstemmed Short-term urban traffic forecasting in smart cities: a dynamic diffusion spatial-temporal graph convolutional network
title_short Short-term urban traffic forecasting in smart cities: a dynamic diffusion spatial-temporal graph convolutional network
title_sort short term urban traffic forecasting in smart cities a dynamic diffusion spatial temporal graph convolutional network
topic Traffic forecasting
Graph convolutional network
Spatial-temporal
Dynamic generation
url https://doi.org/10.1007/s40747-024-01769-6
work_keys_str_mv AT xiangyin shorttermurbantrafficforecastinginsmartcitiesadynamicdiffusionspatialtemporalgraphconvolutionalnetwork
AT junyangyu shorttermurbantrafficforecastinginsmartcitiesadynamicdiffusionspatialtemporalgraphconvolutionalnetwork
AT xiaoyuduan shorttermurbantrafficforecastinginsmartcitiesadynamicdiffusionspatialtemporalgraphconvolutionalnetwork
AT leichen shorttermurbantrafficforecastinginsmartcitiesadynamicdiffusionspatialtemporalgraphconvolutionalnetwork
AT xiaoliliang shorttermurbantrafficforecastinginsmartcitiesadynamicdiffusionspatialtemporalgraphconvolutionalnetwork