Abstract
For complex nonlinear temporal and spatial correlation in traffic flow data, the accurate and effective traffic flow forecasting model is indispensable for understanding the traffic dynamics and predicting the future status of an evolving traffic system. In terms of spatial information extraction, existing approaches are mostly devoted to capture spatial dependency on a predefined graph, which assumes the relation between traffic nodes can be completely offered by an invariant graph structure. However, the fixed graph does not reflect real spatial dependency in traffic data. In this paper, a novel Adaptive Spatial-Temporal Graph Convolutional Network, named as TrafficSCINet, is proposed for traffic flow forecasting. Our model consists of two components: 1) AGCN module uses an adaptive adjacency matrix to dynamically learn the spatial dependencies between traffic nodes under different forecast horizon; 2) SCINet module extracts potential temporal information from traffic flow data through its superb temporal modeling capabilities. Two convolution modules in SCI-Block that have no effect on the results are removed to significantly improve the training speed of the model. Experimental results on four real-world traffic datasets demonstrate that TrafficSCINet achieves state-of-the-art performance consistently than other baselines.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Corey, S., Minh, D.: Streets: A novel camera network dataset for traffic flow. In: Conference and Workshop on Neural Information Processing Systems, pp. 10242–10253, NeurIPS, Vancouver (2019)
Evangelia, C., Christina, I., Christina, M., et al.: Factors affecting bus bunching at the stop level: a geographically weighted regression approach. Int. J. Transport. Sci. Technol. 9(3), 207–217 (2020)
Guo, S.N., Lin, Y.F., Feng, N., et al.: Attention based spatial-temporal graph convolutional networks for traffic flow forecasting. In: AAAI Conference on Artificial Intelligence, pp. 922–929. AAAI, Hawaii (2019)
Li, Y.G., Yu, R., Shahabi, C., et al.: Diffusion convolutional recurrent neural network: Data-driven traffic forecasting. In: International Conference of Learning Representation, pp. 1–16. ICLR, Vancouver (2018)
Mohammed, S.A., Allen, R.C.: Analysis of freeway traffic time-series data by using Box-Jenkins techniques. Transport. Res. Record J. Transport. Res. Board 773(722), 1–9 (1979)
Billy, M.W., Lester, A.H.: Modeling and forecasting vehicular traffic flow as a seasonal Arima process: theoretical basis and empirical results. J. Transp. Eng. 129(6), 664–672 (2003)
Van, L.J., Van, H.C.: Short-term traffic and travel time prediction models. Artif. Intell. Appl. Critical Transp. Issues 22(1), 22–41 (2012)
Jeong, Y.S., Byon, Y.J., Castro-Neto, M.M., et al.: Supervised weighting-online learning algorithm for short-term traffic flow prediction. IEEE Trans. Intell. Transp. Syst. 14(4), 1700–1707 (2013)
Chan, K.Y., Dillon, T., Chang, E., et al.: Prediction of short-term traffic variables using intelligent swarm-based neural networks. IEEE Trans. Control Syst. Technol. 21(1), 263–274 (2013)
Liu, M.H., Zeng, A.L., Chen, M.X., et al.: SCINet: time series modeling and forecasting with sample convolution and interaction. In: Conference and Workshop on Neural Information Processing Systems. NeurIPS, New Orleans (2022)
Wu, Z., Pan, S., Long, G., et al.: Graph wavenet for deep spatial-temporal graph modeling. In: International Joint Conference on Artificial Intelligence. Morgan Kaufmann, Macao (2019)
Bai, L., Yao L.N., Li, C., et al.: Adaptive graph convolutional recurrent network for traffic forecasting. In: Conference and Workshop on Neural Information Processing Systems, pp. 17804–17815. NeurIPS, Online (2020)
Kipf, T.N., Welling, M.: Semi-supervised classification with graph convolutional networks. In: International Conference on Learning Representations. ICLR. Toulon (2017)
Sutskever, I., Vinyals, O., Le, Q.V.: Sequence to sequence learning with neural networks. In: Conference and Workshop on Neural Information Processing Systems, pp. 3104–3112. NeurIPS, Montreal (2014)
Yu, B., Yin, H.T., Zhu, Z.X.: Spatio-temporal graph convolutional networks: a deep learning framework for traffic forecasting. In: International Joint Conferences on Artificial Intelligence. Morgan Kaufmann, Sweden (2017)
Guo, S.N., Lin, Y.F., Feng, N., et al.: Attention based spatial-temporal graph convolutional networks for traffic flow forecasting. In: AAAI Conference on Artificial Intelligence. AAAI, Hawaii (2019)
Song, C., Lin, Y.F., Guo, S.G., et al.: Spatial-temporal synchronous graph convolutional networks: A new framework for spatial-temporal network data forecasting. In: AAAI Conference on Artificial Intelligence. AAAI, New York (2020)
Li, M.Z., Zhu, Z.X.: Spatial-temporal fusion graph neural networks for traffic flow forecasting. In: AAAI Conference on Artificial Intelligence. AAAI, Beijing (2021)
Acknowledgments
This research was funded by the Natural Science Foundation of Shandong Province for Key Project under Grant ZR2020KF006, the National Natural Science Foundation of China under Grant 62273164, the Development Program Project of Youth Innovation Team of Institutions of Higher Learning in Shandong Province, and the Project of Shandong Province Higher Educational Science and Technology Program under Grants J16LB06 and J17KA055.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2023 The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.
About this paper
Cite this paper
Gong, K., Han, S., Yang, X., Yu, W., Guan, Y. (2023). TrafficSCINet: An Adaptive Spatial-Temporal Graph Convolutional Network for Traffic Flow Forecasting. In: Huang, DS., Premaratne, P., Jin, B., Qu, B., Jo, KH., Hussain, A. (eds) Advanced Intelligent Computing Technology and Applications. ICIC 2023. Lecture Notes in Computer Science, vol 14086. Springer, Singapore. https://doi.org/10.1007/978-981-99-4755-3_54
Download citation
DOI: https://doi.org/10.1007/978-981-99-4755-3_54
Published:
Publisher Name: Springer, Singapore
Print ISBN: 978-981-99-4754-6
Online ISBN: 978-981-99-4755-3
eBook Packages: Computer ScienceComputer Science (R0)