Abstract
The complex traffic network spatial correlation and the characteristic of high nonlinear and dynamic traffic conditions in the time are the challenges to accurate traffic flow forecasting. Existing spatiotemporal models attempt to utilize the static graph to explore spatial dependency and employ RNN-based model to capture temporal dependency. However, the static graph fails to reflect the dynamic changeable correlation between each node. That is some nodes have a strong connection in a real traffic network, whereas a weak connection is in a static predefined graph. To overcome the above problems, we propose a spatial dynamic graph convolutional network (SDGCN) for traffic flow forecasting. With the support of an attention fusion network in graph learning, SDGCN generates the dynamic graph at each time step, which can model the changeable spatial correlation from traffic data. By embedding dynamic graph diffusion convolution into gated recurrent unit, our model can explore spatio-temporal dependency simultaneously. Moreover, to handle long sequence forecasting, ReZero transformer is utilized to detect the global temporal correlation capturing. The experiments are conducted on two public datasets. The experimental results demonstrate the superior performance of our network.
Similar content being viewed by others
References
Huakang L u, Ge Z, Song Y, Jiang D, Zhou T, Qin J (2021) A temporal-aware lstm enhanced by loss-switch mechanism for traffic flow forecasting. Neurocomputing 427:169–178
Guo S, Lin Y, Feng N, Song C, Wan H (2019) Attention based spatial-temporal graph convolutional networks for traffic flow forecasting. In: Proceedings of the AAAI conference on artificial intelligence, vol 33, pp 922–929
Wang X, Ma Y, Wang Y, Jin W, Wang X, Tang J, Jia C, Jian Y u (2020) Traffic flow prediction via spatial temporal graph neural network. In: Proceedings of the web conference 2020, pp 1082–1092
Yang S, Li H, Luo Y u, Li J, Song Y, Zhou T (2022) Spatiotemporal adaptive fusion graph network for short-term traffic flow forecasting. Mathematics 10(9):1594
Li Y, Yu R, Cyrus S, Liu Y (2018) Diffusion convolutional recurrent neural network: data-driven traffic forecasting. In: International conference on learning representations
Cui Z, Huang B, Dou H, Cheng Y, Guan J, Zhou T (2022) A two-stage hybrid extreme learning model for short-term traffic flow forecasting. Mathematics 10:2087
Cai L, Zhang Z, Yang J, Yidan Y u, Zhou T, Qin J (2019) A noise-immune kalman filter for short-term traffic flow forecasting. Phys: Stat Mech Appl 536:122601
Zhou T, Jiang D, Lin Z, Han G, Xuemiao X u, Qin J (2019) Hybrid dual kalman filtering model for short-term traffic flow forecasting. IET Intell Transp Syst 13(6):1023–1032
Yu H, Wu Z, Wang S, Wang Y, Ma X (2017) Spatiotemporal recurrent convolutional networks for traffic prediction in transportation networks. Sensors 17(7):1501
Du B, Peng H, Wang S, Md ZAB, Wang L, Gong Q, Liu L, Li J (2019) Deep irregular convolutional residual lstm for urban traffic passenger flows prediction. IEEE Trans Intell Transp Syst 21(3):972–985
Cai L, Lei M, Zhang S, Yidan Y u, Zhou T, Qin J (2020) A noise-immune lstm network for short-term traffic flow forecasting. Chaos 30(2):023135
Zhang X, Huang C, Yong X u, Xia L, Dai P, Bo L, Zhang J, Zheng Y u (2021) Traffic flow forecasting with spatial-temporal graph diffusion network. In: Proceedings of the AAAI conference on artificial intelligence, vol 35, pp 15008–15015
Sun K, Zhu Z, Lin Z (2020) Adagcn: adaboosting graph convolutional networks into deep models. In: International conference on learning representations
Wu Z, Pan S, Long G, Jing J, Zhang C (2019) Graph wavenet for deep spatial-temporal graph modeling. In: Proceedings of the 28th international joint conference on artificial lntelligence, IJCAI ’19. AAAI Press, pp 1907–1913
Zonghan W u, Pan S, Chen F, Long G, Zhang C, Philip S Y u (2020) A comprehensive survey on graph neural networks. IEEE Trans Neural Netw Learning Syst 32(1):4–24
Li M, Zhu Z (2021) Spatial-temporal fusion graph neural networks for traffic flow forecasting. In: Proceedings of the AAAI conference on artificial intelligence, vol 35, pp 4189–4196
Zheng C, Fan X, Wang C, Jianzhong Qi (2020) Gman: a graph multi-attention network for traffic prediction. In: Proceedings of the AAAI conference on artificial intelligence, vol 34, pp 1234–1241
Chen C, Li K, Teo SG, Zou X, Wang K, Wang J, Zeng Z (2019) Gated residual recurrent graph neural networks for traffic prediction. In: Proceedings of the AAAI conference on artificial intelligence, vol 33, pp 485–492
Bing Y u, Yin H, Zhanxing Zhu (2018) Spatio-temporal graph convolutional networks: a deep learning framework for traffic forecasting. IJCAI’18. AAAI Press, pp 3634–3640
Zhang X, Haghani A, Yang S (2019) Is dynamic traffic sensor network profitable for network-level real-time information prediction? Transport Res Part C: Emerging Technol 102:32–59
Hochreiter S, Bengio Y, Frasconi P, Schmidhuber J et al (2001) Gradient flow in recurrent nets: the difficulty of learning long-term dependencies
Li F, Feng J, Yan H, Jin G, Yang F, Sun F, Jin D, Li Y (2021) Dynamic graph convolutional recurrent network for traffic prediction: benchmark and solution. ACM Trans Knowl Discovery Data (TKDD)
Zhao L, Song Y, Zhang C, Liu Y u, Wang P u, Lin T, Deng M, Li H (2019) T-gcn: a temporal graph convolutional network for traffic prediction. IEEE Trans Intell Transp Syst 21(9):3848–3858
Guo S, Lin Y, Wan H, Li X, Cong G (2021) Learning dynamics and heterogeneity of spatial-temporal graph data for traffic forecasting. IEEE Trans Knowl Data Eng
Bachlechner T, Majumder BP, Mao H, Cottrell G, McAuley J (2021) Rezero is all you need: fast convergence at large depth. In: Uncertainty in artificial intelligence. PMLR, pp 1352–1361
Chiang Wei-Lin, Liu X, Si S i, Li Y, Bengio S, Hsieh C-J (2019) Cluster-gcn: an efficient algorithm for training deep and large graph convolutional networks. In: Inproceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining, pp 257–266
Wang J, Song G, Yi W u, Wang L (2020) Streaming graph neural networks via continual learning. In: Proceedings of the 29th ACM international conference on information & knowledge management, pp 1515–1524
Wang C, Qiu Y, Gao D, Scherer S (2022) Lifelong graph learning. In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pp 13719–13728
Welling M, Kipf TN (2016) Semi-supervised classification with graph convolutional networks. In: J. international conference on learning representations (ICLR 2017)
Huakang L u, Huang D, Youyi S, Jiang D, Zhou T, Jing Qin (2020) St-trafficnet: a spatial-temporal deep learning network for traffic forecasting. Electronics 9(9):1–17
Kong X, Zhang J, Wei X, Xing W, Wei L u (2022) Adaptive spatial-temporal graph attention networks for traffic flow forecasting. Appl Intell 52(4):4300–4316
Young T, Hazarika D, Poria S, Cambria E (2018) Recent trends in deep learning based natural language processing. IEEE ComputIntell Magazine 13(3):55–75
Liu G, Guo J (2019) Bidirectional lstm with attention mechanism and convolutional layer for text classification. Neurocomputing 337:325–338
Zakir Hossain MD, Sohel F, Shiratuddin MF, Laga H (2019) A comprehensive survey of deep learning for image captioning. ACM Computing Surveys (CsUR) 51(6):1–36
Huang B, Tan G, Song Y, Zhou T, Dou H, Cui Z (2022) Mutual gain adaptive network for segmenting brain stroke lesions. Appl Soft Comput
Zhou T, Dou H, Tan J, Song Y, Wang F, Wang J (2022) Small dataset solves big problem: an outlier-insensitive binary classifier for inhibitory potency prediction knowledge-based systems
Dou H, Tan J, Wei H, Wang F, Yang J, Ma X-G, Wang J, Teng Z (2022) Transfer inhibitory potency prediction to binary classification: a model only needs a small training set computer methods and programs in biomedicine
Zheng L, Guo N, Chen W, Jin Y u, Jiang D (2020) Sentiment-guided sequential recommendation. In: Proceedings of the 43rd international ACM SIGIR conference on research and development in information retrieval, pp 1957–1960
Fang W, Zhuo W, Yan J, Song Y, Jiang D, memory Teng Zhou. (2022) Attention meets long short-term a deep learning network for traffic flow forecasting. Physica A: Stat Mech Appl 587:126485
Huang R, Huang C, Liu Y, Dai G, Kong W (2020) Lsgcn: long short-term traffic prediction with graph convolutional networks. IJCAI, pp 2355–2361
Vaswani A, Shazeer N, Parmar N, Uszkoreit J, Jones L, Gomez AN, Kaiser Łukasz, Polosukhin I (2017) Attention is all you need. In: Advances in neural information processing systems, pp 5998–6008
Zonghan W u, Pan S, Long G, Jiang J, Chang X, Zhang Chengqi (2020) Connecting the dots Multivariate time series forecasting with graph neural networks. In: Proceedings of the 26th ACM SIGKDD international conference on knowledge discovery & data mining, pp 753–763
Lippi M, Bertini M, forecasting Paolo Frasconi. (2013) Short-term traffic flow an experimental comparison of time-series analysis and supervised learning. IEEE Trans Intell Transp Syst 14(2):871–882
Keskar NS, Mudigere D, Nocedal J, Smelyanskiy M, Tang PTP (2017) On large-batch training for deep learning: generalization gap and sharp minima. In: 5th international conference on learning representations, ICLR 2017, Toulon, France, 24-26, April 2017, conference track proceedings. OpenReview.net
Smith LN (2017) Cyclical learning rates for training neural networks. In: 2017 IEEE Winter conference on applications of computer vision (WACV). IEEE, pages 464–472
Bai Lei, Yao Lina, Li Can, Wang Xianzhi, Wang Can (2020) Adaptive graph convolutional recurrent network for traffic forecasting. In: Advances in neural information processing systems 33: annual conference on neural information processing systems 2020, NeurIPS 2020, 6-12, December 2020, virtual
Acknowledgements
The data that support the findings of this study are available from the corresponding author upon reasonable request. Huaying Li and Shumin Yang contribute equally. This work was supported by the National Natural Science Foundation of China (No. 61902232), the 2022 Guangdong Basic and Applied Basic Research Foundation (No. 2022A1515011590), the STU Incubation Project for the Research of Digital Humanities and New Liberal Arts (No. 2021DH-3), the 2020 Li Ka Shing Foundation Cross-Disciplinary Research Grant (No. 2020LKSFG05D), the Innovation School Project of Guangdong Province (No. 2017KCXTD015), and the Open Fund of Guangdong Provincial Key Laboratory of Infectious Diseases and Molecular Immunopathology (No. GDKL202212).
Author information
Authors and Affiliations
Corresponding author
Additional information
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Huaying Li and Shumin Yang contribute equally.
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Li, H., Yang, S., Song, Y. et al. Spatial dynamic graph convolutional network for traffic flow forecasting. Appl Intell 53, 14986–14998 (2023). https://doi.org/10.1007/s10489-022-04271-z
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10489-022-04271-z