Skip to main content

Advertisement

Springer Nature Link
Log in

Spatial-temporal dynamic semantic graph neural network

  • Original Article
  • Published:
Neural Computing and Applications Aims and scope Submit manuscript

Abstract

Most existing methods based on graph neural network for traffic flow forecasting cannot effectively exploit potential semantic features, multiple features are aggregated without refining the proportion of their respective weights, and the over-smoothing problem limits the stacked depth of the model. To solve these problems, spatial-temporal dynamic semantic graph neural network is proposed in this paper. Firstly, two different semantic features have been captured by dynamic time warping algorithm and Pearson correlation coefficient for constructing two semantic adjacency matrices. Secondly, a dynamic aggregation method is proposed that learns the weighting ratio corresponding to each feature through training. Thirdly, the injection-stacked structure is designed to solve the over-smoothing problem and allow the network to be stacked with more layer and improve the forecasting accuracy. Finally, the experiments on four PEMS datasets with various methods such as spatio-temporal graph convolutional networks, attention-based spatial-temporal graph convolutional networks, etc. verify that spatial-temporal dynamic semantic graph neural network obtains minimal forecasting errors by capturing the potential semantic features, dynamically aggregating multiple features, and deepening the network layers by injecting-stacked structure. It achieves that root mean square error is 25.59, mean absolute error is 16.12 and mean absolute percentage error is 16.15 on the PEMS03 dataset.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Subscribe and save

Springer+ Basic
$34.99 /Month
  • Get 10 units per month
  • Download Article/Chapter or eBook
  • 1 Unit = 1 Article or 1 Chapter
  • Cancel anytime
Subscribe now

Buy Now

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7
Fig. 8
Fig. 9
Fig. 10

Similar content being viewed by others

Explore related subjects

Discover the latest articles, news and stories from top researchers in related subjects.

References

  1. Bai L, Yao L, Kanhere S, Wang X., Sheng Q et al (2019) Stg2seq: Spatial-temporal graph to sequence model for multi-step passenger demand forecasting. arXiv preprint arXiv:1905.10069

  2. Yu B, Yin H, Zhu Z (2017) Spatio-temporal graph convolutional networks: A deep learning framework for traffic forecasting. arXiv preprint arXiv:1709.04875

  3. Wu Z, Pan S, Long G, Jiang J, Zhang C (2019) Graph wavenet for deep spatial-temporal graph modeling. arXiv preprint arXiv:1906.00121

  4. 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

  5. Gilmer J, Schoenholz SS, Riley PF, Vinyals O, Dahl GE (2017) Neural message passing for quantum chemistry. In: International conference on machine learning, pp 1263–1272. PMLR

  6. Li Q, Han Z, Wu X-M (2018) Deeper insights into graph convolutional networks for semi-supervised learning. In: Thirty-second AAAI conference on artificial intelligence

  7. Yule GU (1927) Vii. On a method of investigating periodicities disturbed series, with special reference to wolfer’s sunspot numbers. Philos Trans R Soc Lond Ser A, Containing Papers of a Mathematical or Physical Character 226(636–646):267–298

  8. Enders W (2004) Applied econometric time series, 2nd edn. Willey, Hoboken

    Google Scholar 

  9. Hamilton JD (2020) Time series analysis. Princeton University Press, Princeton

    Book  Google Scholar 

  10. Drucker H, Burges CJ, Kaufman L, Smola A, Vapnik V (1997) Support vector regression machines. Adv Neural Inf Process Syst 9:155–161

    Google Scholar 

  11. Beutler FJ (1965) Prediction and regulation by linear least-squares methods. JSTOR

  12. Box GE, Pierce DA (1970) Distribution of residual autocorrelations in autoregressive-integrated moving average time series models. J Am Stat Assoc 65(332):1509–1526

    Article  MathSciNet  Google Scholar 

  13. Kipf TN, Welling M (2016) Semi-supervised classification with graph convolutional networks. arXiv preprint arXiv:1609.02907

  14. Li Y, Yu R, Shahabi C, Liu Y (2017) Diffusion convolutional recurrent neural network: data-driven traffic forecasting. arXiv preprint arXiv:1707.01926

  15. Rumelhart DE, Hinton GE, Williams RJ (1986) Learning representations by back-propagating errors. Nature 323(6088):533–536

    Article  Google Scholar 

  16. Hochreiter S, Schmidhuber J (1997) Long short-term memory. Neural Comput 9(8):1735–1780

    Article  Google Scholar 

  17. Wang X, Chen C, Min Y, He J, Yang B, Zhang Y (2018) Efficient metropolitan traffic prediction based on graph recurrent neural network. arXiv preprint arXiv:1811.00740

  18. Prokoptsev NG, Alekseenko A, Kholodov YA (2018) Traffic flow speed prediction on transportation graph with convolutional neural networks. Comput Res Model 10(3):359–367

    Article  Google Scholar 

  19. Bai S, Kolter JZ, Koltun V (2018) An empirical evaluation of generic convolutional and recurrent networks for sequence modeling. arXiv preprint arXiv:1803.01271

  20. Chen C, Petty K, Skabardonis A, Varaiya P, Jia Z (2001) Freeway performance measurement system: mining loop detector data. Transp Res Rec 1748(1):96–102

    Article  Google Scholar 

  21. Cui Z, Henrickson K, Ke R, Wang Y (2019) Traffic graph convolutional recurrent neural network: a deep learning framework for network-scale traffic learning and forecasting. IEEE Trans Intell Transp Syst 21(11):4883–4894

    Article  Google Scholar 

  22. Yan B, Wang G, Yu J, Jin X, Zhang H (2021) Spatial-temporal chebyshev graph neural network for traffic flow prediction in iot-based its. IEEE Internet of Things J

  23. Xie Y, Xiong Y, Zhu Y (2020) Sast-gnn: A self-attention based spatio-temporal graph neural network for traffic prediction. In: International conference on database systems for advanced applications. Springer, pp 707–714

  24. Agafonov A (2020) Traffic flow prediction using graph convolution neural networks. In: 2020 10th international conference on information science and technology (ICIST). IEEE, pp 91–95

  25. Du S, Li T, Gong X, Horng S-J (2018) A hybrid method for traffic flow forecasting using multimodal deep learning. arXiv preprint arXiv:1803.02099

Download references

Acknowledgements

This work was partially supported by the National Natural Science Foundation of China (Grant Nos. 41974033 and 61803208), the Scientific and technological achievements transformation project of Jiangsu Province (BA2020004), 2020 Industrial Transformation and Upgrading Project of Industry and Information Technology Department of Jiangsu Province, Postgraduate Research & Practice Innovation Program of Jiangsu Province, Bidding project for breakthroughs in key technologies of advantageous industries in Nanjing (2018003).

Author information

Authors and Affiliations

  1. School of Electrical and Automation Engineering, Nanjing Normal University, Xuelin Road 2, Nanjing, 210023, Jiangsu, China

    Rui Zhang & Fei Xie

  2. College of Civil Aviation, Nanjing University of Aeronautics and Astronautics, Jiangjun Street 29, Nanjing, 211106, Jiangsu, China

    Rui Sun

  3. Center for Transport Studies, Imperial College London, London, SW7 2AZ, UK

    Rui Sun

  4. College of Mechanical and Electronic Engineering, Nanjing Forestry University, Longpandong Road 159, Nanjing, 210037, Jiangsu, China

    Lei Huang

  5. School of Instrument Science and Engineering, Southeast University, Sipailou 2, Nanjing, 210096, Jiangsu, China

    Xixiang Liu

  6. School of Instrument Science and Engineering, Nanjing Institute of Technology, Hongjing Avenue 1, Nanjing, 211167, Jiangsu, China

    Jianjun Shi

  7. Nanjing Zhongke Raycham Laser Technology Co., Ltd, Xinghe Road 18, Nanjing, 210038, Jiangsu, China

    Jianjun Shi

Authors
  1. Rui Zhang
    View author publications

    You can also search for this author inPubMed Google Scholar

  2. Fei Xie
    View author publications

    You can also search for this author inPubMed Google Scholar

  3. Rui Sun
    View author publications

    You can also search for this author inPubMed Google Scholar

  4. Lei Huang
    View author publications

    You can also search for this author inPubMed Google Scholar

  5. Xixiang Liu
    View author publications

    You can also search for this author inPubMed Google Scholar

  6. Jianjun Shi
    View author publications

    You can also search for this author inPubMed Google Scholar

Corresponding author

Correspondence to Fei Xie.

Ethics declarations

Conflict of interest

The authors declare that they have no conflict of interest.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Zhang, R., Xie, F., Sun, R. et al. Spatial-temporal dynamic semantic graph neural network. Neural Comput & Applic 34, 16655–16668 (2022). https://doi.org/10.1007/s00521-022-07285-3

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s00521-022-07285-3

Keywords