OpenLB
Open Library of Bioscience
Advanced Search
Scientific reports
Mar 15, 2025;15 (1): 8935.

Graph convolution networks based on adaptive spatiotemporal attention for traffic flow forecasting.

Hongbo Xiao, Beiji Zou, Jianhua Xiao
Author Information
  1. Hongbo Xiao: School of Computer Science and Engineering, Central South University, Changsha, 410083, China.
  2. Beiji Zou: School of Computer Science and Engineering, Central South University, Changsha, 410083, China. bjzou@csu.edu.cn.
  3. Jianhua Xiao: School of Computer and Artificial Intelligence (School of Software), Huaihua University, Huaihua, 418000, China. xiaojianhua@hnu.edu.cn.
PMID: 40089544 DOI: 10.1038/s41598-025-88706-w

Abstract

Traffic flow is the most direct indicator of traffic conditions, and accurate prediction of traffic flow is a key challenge for scholars in the field of intelligent transportation. However, traffic flow displays significant nonlinearity, dynamic changes, spatiotemporal dependencies, and most existing methods overlook the influence of road topology on the spatiotemporal properties of traffic flow, creating substantial challenges for traffic flow prediction. This paper proposes a graph convolutional traffic flow prediction model based on adaptive spatiotemporal attention. Initially, the model adaptively adjusts spatiotemporal weight distribution using a meticulously designed spatiotemporal attention mechanism, effectively capturing dynamic spatiotemporal correlations in traffic data. Subsequently, it integrates graph convolutional neural networks (GCNs) with long short-term memory (LSTM) networks to capture the spatiotemporal characteristics of traffic data. Additionally, a GCN is designed to capture the spatial topological relationships of the road network. Finally, a novel fusion mechanism is introduced to integrate the spatiotemporal features of traffic data with the spatial topological relationships of roads, aiming to achieve accurate predictions. Experimental results demonstrate that the model proposed in this paper outperforms six selected baseline methods.

Keywords

GCN LSTM Spatiotemporal Attention Mechanism Traffic Flow Prediction

References

  1. Qi, X., Mei, G. & Piccialli, X. F. A deep learning approach for long-term traffic flow prediction with multifactor fusion using spatiotemporal graph convolutional network. IEEE transactions on intelligent transportation systems 24, 8687���8700 (2023). [DOI: 10.1109/TITS.2022.3201879]
  2. Chen, Y. et al. Graph attention network with spatial-temporal clustering for traffic flow forecasting in intelligent transportation system. IEEE Transactions on Intelligent Transportation Systems 24, 8727���8737. https://doi.org/10.1109/TITS.2022.3208952 (2023). [DOI: 10.1109/TITS.2022.3208952]
  3. Luo, Q., He, S., Han, X., Wang, Y. & Li, H. Lsttn: A long-short term transformer-based spatiotemporal neural network for traffic flow forecasting. Knowledge-Based Systems 293 (2024).
  4. Li, Q., Yang, X., Wang, Y., Wu, Y. & He, D. Spatial-temporal traffic modeling with a fusion graph reconstructed by tensor decomposition. IEEE transactions on intelligent transportation systems 25 (2024).
  5. Msahli, M., Qiu, H., Zheng, Q., Memmi, G. & Lu, J. Topological graph convolutional network-based urban traffic flow and density prediction. IEEE Transactions on Intelligent Transportation Systems PP (2020).
  6. Williams, B. M. & Hoel, L. A. Modeling and forecasting vehicular traffic flow as a seasonal arima process: Theoretical basis and empirical results. Journal of Transportation Engineering 129, 664���672 (2003). [DOI: 10.1061/(ASCE)0733-947X(2003)129]
  7. Lu, Z., Zhou, C., Wu, J., Jiang, H. & Cui, S. Integrating granger causality and vector auto-regression for traffic prediction of large-scale wlans. Ksii Transactions on Internet & Information Systems 10, 136���151 (2016).
  8. Chen, T. & Lu, S. Accurate and efficient traffic sign detection using discriminative adaboost and support vector regression. IEEE Transactions on Vehicular Technology 65, 4006���4015 (2016). [DOI: 10.1109/TVT.2015.2500275]
  9. Zuduo, Zheng, Dongcai & Su. Short-term traffic volume forecasting: A k-nearest neighbor approach enhanced by constrained linearly sewing principle component algorithm. Transportation Research Part C Emerging Technologies 43, 143���157 (2014).
  10. Cheng, X., Zhang, R., Zhou, J. & Xu, W. Deeptransport: Learning spatial-temporal dependency for traffic condition forecasting. In 2018 International Joint Conference on Neural Networks (IJCNN), 1���8 (2018).
  11. Defferrard, M., Bresson, X. & Vandergheynst, P. Convolutional neural networks on graphs with fast localized spectral filtering. In Neural Information Processing Systems (2016).
  12. Chondrodima, E., Pelekis, N., Pikrakis, A. & Theodoridis, Y. An efficient lstm neural network-based framework for vessel location forecasting. IEEE transactions on intelligent transportation systems 24 (2023).
  13. Ahmed, M. S. & Cook, A. R. Analysis of freeway traffic time-series data by using box-jenkins techniques. Transportation Research Record (1979).
  14. Yu, J. & Zuo, M. A novel traffic flow detection method using multiple statistical parameters. In International Conference on Measuring Technology and Mechatronics Automation, 51���54 (2015).
  15. Zhu, Z., Xu, M., Ke, J., Yang, H. & Chen, X. M. A bayesian clustering ensemble gaussian process model for network-wide traffic flow clustering and prediction. Transportation Research Part C: Emerging Technologies 148, 104032��� (2023).
  16. Zhang, L., Liu, Q., Yang, W., Wei, N. & Dong, D. An improved k-nearest neighbor model for short-term traffic flow prediction. Procedia - Social and Behavioral Sciences 96, 653���662 (2013). [DOI: 10.1016/j.sbspro.2013.08.076]
  17. Lippi, M., Bertini, Frasconi & P. Short-term traffic flow forecasting: An experimental comparison of time-series analysis and supervised learning. IEEE Transactions on Intelligent Transportation Systems 14, 653���662 (2013).
  18. Tao, Q., Li, Z. & Suykens, J. X. L. D. S. A. K. Short-term traffic flow prediction based on the efficient hinging hyperplanes neural network. IEEE transactions on intelligent transportation systems 23, 15616���15628 (2022).
  19. Liu, H. et al. Staeformer: spatio-temporal adaptive embedding makes vanilla transformer sota for traffic forecasting. In Proceedings of the 32nd ACM International Conference on Information and Knowledge Management (CIKM), Birmingham, UK, 21���25 (2023).
  20. Qin, J. et al. Muse-net: Disentangling multi-periodicity for traffic flow forecasting. In 2024 IEEE 40th International Conference on Data Engineering (ICDE), 1282���1295 (IEEE, 2024).
  21. Zhou, Y. et al. Make graph neural networks great again: A generic integration paradigm of topology-free patterns for traffic speed prediction. arXiv preprint[SPACE] arXiv:2406.16992 (2024).
  22. Ren, Y., Zhao, D., Luo, D., Ma, H. & Duan, P. Global-local temporal convolutional network for traffic flow prediction. IEEE Transactions on Intelligent Transportation Systems PP, 1���7 (2020).
  23. Zhang, Y. et al. Modeling spatio-temporal mobility across data silos via personalized federated learning. IEEE Transactions on Mobile Computing (2024).
  24. Bai, L., Yao, L., Li, C., Wang, X. & Wang, C. Adaptive graph convolutional recurrent network for traffic forecasting. ArXiv abs/2007.02842 (2020).
  25. Tom, Y., Devamanyu, H., Soujanya, P. & Erik, C. Recent trends in deep learning based natural language processing [review article]. IEEE Computational Intelligence Magazine 13, 55���75 (2018). [DOI: 10.1109/MCI.2018.2840738]
  26. Wang, L., Cao, Z., de Melo, G. & Liu, Z. Relation classification via multi-level attention cnns. In Annual Meeting of the Association for Computational Linguistics, 1298���1307 (2016).
  27. Cinar, Y. G. et al. Position-based content attention for time series forecasting with sequence-to-sequence rnns. In Neural Information Processing, 533���544 (2017).
  28. Huang, X., Tang, J., Yang, X. & Xiong, L. A time-dependent attention convolutional lstm method for traffic flow prediction. Applied Intelligence 52, 17371���17386 (2022). [DOI: 10.1007/s10489-022-03324-7]
  29. Chauhan, N. S., Kumar, N. & Eskandarian, A. A novel confined attention mechanism driven bi-gru model for traffic flow prediction. IEEE Transactions on Intelligent Transportation Systems 25, 9181���9191. https://doi.org/10.1109/TITS.2024.3375890 (2024). [DOI: 10.1109/TITS.2024.3375890]
  30. Fang, W. et al. Attention meets long short-term memory: A deep learning network for traffic flow forecasting. Physica A: Statistical Mechanics and its Applications 587, S0378437121007585 (2022). [DOI: 10.1016/j.physa.2021.126485]
  31. Zhao, J. et al. Adaptive graph convolutional network-based short-term passenger flow prediction for metro. Journal of Intelligent Transportation Systems (2023).
  32. Cui, Z., Henrickson, K., Ke, R. & Wang, Y. Traffic graph convolutional recurrent neural network: A deep learning framework for network-scale traffic learning and forecasting. IEEE Transactions on Intelligent Transportation Systems 21, 4883���4894. https://doi.org/10.1109/TITS.2019.2950416 (2020). [DOI: 10.1109/TITS.2019.2950416]
  33. Agafonov, A. Traffic flow prediction using graph convolution neural networks. In 2020 10th International Conference on Information Science and Technology (ICIST), 91���95, https://doi.org/10.1109/ICIST49303.2020.9201971 (2020).
  34. Yu, B., Yin, H. & Zhu, Z. Spatio-temporal graph convolutional networks: a deep learning framework for traffic forecasting. In International Joint Conference on Artificial Intelligence, 3634 ��� 3640 (2018).
  35. Xia, M., Jin, D. & Chen, J. Short-term traffic flow prediction based on graph convolutional networks and federated learning. IEEE Transactions on Intelligent Transportation Systems 24, 1191���1203. https://doi.org/10.1109/TITS.2022.3179391 (2023). [DOI: 10.1109/TITS.2022.3179391]
  36. Lv, M. et al. Temporal multi-graph convolutional network for traffic flow prediction. IEEE Transactions on Intelligent Transportation Systems 22, 3337���3348. https://doi.org/10.1109/TITS.2020.2983763 (2021). [DOI: 10.1109/TITS.2020.2983763]
  37. Zhang, W. et al. Irregular traffic time series forecasting based on asynchronous spatio-temporal graph convolutional networks. In Proceedings of the 30th ACM SIGKDD Conference on Knowledge Discovery and Data Mining, 4302���4313 (2024).
  38. Zhang, Y. et al. Adaptive and interactive multi-level spatio-temporal network for traffic forecasting. IEEE Transactions on Intelligent Transportation Systems (2024).
  39. Weng, W. et al. A decomposition dynamic graph convolutional recurrent network for traffic forecasting. Pattern Recognition 142, 109670 (2023). [DOI: 10.1016/j.patcog.2023.109670]
  40. Kipf, T. N. & Welling, M. Semi-Supervised Classification with Graph Convolutional Networks. arXiv e-prints[SPACE] arXiv:1609.02907 , https://doi.org/10.48550/arXiv.1609.02907 (2016).
  41. Chen, C. & Varaiya, P. P. Freeway performance measurement system (pems). PATH research report 39, 67���75 (2002).
  42. Chen, et al. A combined model based on seasonal autoregressive integrated moving average and modified particle swarm optimization algorithm for electrical load forecasting. Journal of intelligent & fuzzy systems: Applications in Engineering and Technology 32, 3447���3459 (2017). [DOI: 10.3233/JIFS-169283]
  43. Yu, Q. & Li, Z. Correlated load forecasting in active distribution networks using spatial-temporal synchronous graph convolutional networks. IET Energy Systems Integration 3, 355���366 (2021). [DOI: 10.1049/esi2.12028]

Grants

  1. 62202198/the National Natural Science Foundation of China
  2. 62202198/the National Natural Science Foundation of China
Journal Article

Word Cloud

Created with Highcharts 10.0.0trafficspatiotemporalflowpredictionmodelattentiondatanetworksTrafficaccuratedynamicmethodsroadpapergraphconvolutionalbasedadaptivedesignedmechanismLSTMcaptureGCNspatialtopologicalrelationshipsdirectindicatorconditionskeychallengescholarsfieldintelligenttransportationHoweverdisplayssignificantnonlinearitychangesdependenciesexistingoverlookinfluencetopologypropertiescreatingsubstantialchallengesproposesInitiallyadaptivelyadjustsweightdistributionusingmeticulouslyeffectivelycapturingcorrelationsSubsequentlyintegratesneuralGCNslongshort-termmemorycharacteristicsAdditionallynetworkFinallynovelfusionintroducedintegratefeaturesroadsaimingachievepredictionsExperimentalresultsdemonstrateproposedoutperformssixselectedbaselineGraphconvolutionforecastingSpatiotemporalAttentionMechanismFlowPrediction

Similar Articles

Cited By