Smart City Gnosys

Smart city article details

Title	Dyn-Gwn: Application Of Graph Wave Networks On The Largest Traffic Dataset
ID_Doc	21215
Authors	Wang J.; He R.; He Y.; Yang B.; Zhou R.; Xu J.; Peng J.
Year	2024
Published	Proceedings - 2024 International Symposium on Internet of Things and Smart Cities, ISITSC 2024
DOI	http://dx.doi.org/10.1109/ISITSC64373.2024.00020
Abstract	Urban road traffic prediction is one of the key technologies in the construction of smart cities, playing a crucial role in alleviating traffic congestion and optimizing traffic management. Historically, high-quality and large-scale benchmark datasets have proven invaluable in driving research forward. For instance, the emerging LargeST dataset, as a representative large-scale dataset, has been evaluated by 12 different models, some of which, such as D2STGNN, GWNET, and STGCN, have shown excellent performance. This raises an important question: Is there a model that can achieve better performance on large-scale datasets like LargeST? This paper introduces a novel model recently proposed-DynGWN, specifically designed for time series prediction and spatio-temporal data analysis, particularly suitable for traffic prediction scenarios. DynGWN integrates the advantages of Graph Neural Networks (GNN) and Convolutional Neural Networks (CNN), dynamically learning the graph structure to more effectively capture spatio-temporal dependencies. Additionally, the model proposes an effective method for automatically learning hidden spatial dependencies from data. Given the potential research value of DynGWN for the performance on large-scale datasets in future work, this paper applies DynGWN to the LargeST large-scale dataset and compares it with 12 baseline models that have been evaluated on this dataset. The results show that DynGWN has a slight edge in road traffic prediction, out-performing not only traditional static graph models but also the 12 baseline models. This proves that DynGWN not only performs well on small-scale datasets but also has excellent performance on large-scale datasets. At the same time, this paper also points out the direction for future research and the current shortcomings. The disadvantage is that limited by experimental conditions, there are only 5000 training samples in a single time, and the batch size is only 8. In the future, then, we will continue to explore which parameter configurations will make the model perform better on large traffic datasets. We release and baseline implementations at https://github.com/hyj2004/Dyn-GWN-Application-of-Graph-Wave-Networks-on-the-Largest-Traffic-Dataset. © 2024 IEEE.
Author Keywords	Baseline model; DynGWN; Graph WaveNet; large-scale dataset

Similar Articles

Id	Similarity	Authors	Title	Published
20800	0.911	Diao Z.; Wang X.; Zhang D.; Xie G.; Chen J.; Pei C.; Meng X.; Xie K.; Zhang G.	Dmstg: Dynamic Multiview Spatio-Temporal Networks For Traffic Forecasting	IEEE Transactions on Mobile Computing, 23, 6 (2024)
1982	0.91	Sharma A.; Sharma A.; Nikashina P.; Gavrilenko V.; Tselykh A.; Bozhenyuk A.; Masud M.; Meshref H.	A Graph Neural Network (Gnn)-Based Approach For Real-Time Estimation Of Traffic Speed In Sustainable Smart Cities	Sustainability (Switzerland), 15, 15 (2023)
58677	0.906	Lian P.; Li Y.; Liu B.; Feng X.	Traffic Speed Prediction Using Multivariate Time Series Dynamic Graph Neural Network; [基于多元时间序列动态图神经网络的交通速度预测]	Journal of Geo-Information Science, 27, 3 (2025)
21283	0.905	Li F.; Feng J.; Yan H.; Jin G.; Yang F.; Sun F.; Jin D.; Li Y.	Dynamic Graph Convolutional Recurrent Network For Traffic Prediction: Benchmark And Solution	ACM Transactions on Knowledge Discovery from Data, 17, 1 (2023)
28556	0.903	Jinia S.N.; Azad S.B.; Akter R.; Dipto T.R.; Khaliluzzaman M.	Gtn-Gcn: Real-Time Traffic Forecasting Using Graph Convolutional Network And Transformer	Applied Computational Intelligence and Soft Computing, 2025, 1 (2025)
21123	0.902	Hu J.; Lin X.; Wang C.	Dstgcn: Dynamic Spatial-Temporal Graph Convolutional Network For Traffic Prediction	IEEE Sensors Journal, 22, 13 (2022)
52507	0.897	Liu S.; Zhu J.; Lei W.; Zhang P.	Spatial-Temporal Attention Graph Wavenet For Traffic Forecasting	2023 5th International Conference on Data-Driven Optimization of Complex Systems, DOCS 2023 (2023)
37214	0.893	Yu W.; Wu S.; Huang M.	Mmgfra: A Multiscale Multigraph Learning Framework For Traffic Prediction In Smart Cities	Earth Science Informatics, 16, 3 (2023)
38024	0.892	Yao H.; Chen R.; Xie Z.; Yang J.; Hu M.; Guo J.	Mra-Dgcn: Multi-Range Attention-Based Dynamic Graph Convolutional Network For Traffic Prediction	Proceedings - 2022 IEEE International Conference on Big Data, Big Data 2022 (2022)
52568	0.891	Ren H.; Kang J.; Zhang K.	Spatio-Temporal Graph-Tcn Neural Network For Traffic Flow Prediction	2022 19th International Computer Conference on Wavelet Active Media Technology and Information Processing, ICCWAMTIP 2022 (2022)