Multi-Scale Dynamic Hypergraph Convolutional Network for Traffic Flow Forecasting

doi:10.1007/s12204-023-2682-z

Abstract

Abstract: This paper focuses on the problem of traffic flow forecasting, with the aim of forecasting future traffic conditions based on historical traffic data. This problem is typically tackled by utilizing spatio-temporal graph neural networks to model the intricate spatio-temporal correlations among traffic data. Although these methods have achieved performance improvements, they often suffer from the following limitations: These methods face challenges in modeling high-order correlations between nodes. These methods overlook the interactions between nodes at different scales. To tackle these issues, in this paper, we propose a novel model named multi-scale dynamic hypergraph convolutional network (MSDHGCN) for traffic flow forecasting. Our MSDHGCN can effectively model the dynamic higher-order relationships between nodes at multiple time scales, thereby enhancing the capability for traffic forecasting. Experiments on two real-world datasets demonstrate the effectiveness of the proposed method.

Key words: traffic flow forecasting, dynamic hypergraph, hypergraph structure learning, multi-time scale

摘要： 本文关注交通流量预测问题，旨在基于历史交通数据预测未来的交通状况。通常情况下，通过利用时空图神经网络建模交通数据之间复杂的时空相关性来解决这一问题。尽管这些方法在性能上有所提升，但往往存在以下限制：这些方法难以建模节点之间的高阶相关性；这些方法忽视了不同尺度下节点之间的相互作用。为了解决这些问题，本文提出了面向交通流量预测的多尺度动态超图卷积网络（MSDHGCN）。提出的MSDHGCN能够有效地建模多个时间尺度上节点之间的动态高阶关系，从而增强交通预测的能力。在两个真实数据集上的实验证明了所提出方法的有效性。

关键词: 交通流量预测，动态超图，超图结构学习，多时间尺度

CLC Number:

TP183

DONG Zhaoxian, YU Shuo, SHEN Yanming. Multi-Scale Dynamic Hypergraph Convolutional Network for Traffic Flow Forecasting[J]. J Shanghai Jiaotong Univ Sci, 2025, 30(5): 880-888.

References

[1] ZHANG J P, WANG F Y, WANG K F, et al. Data-driven intelligent transportation systems: A survey [J]. IEEE Transactions on Intelligent Transportation Systems, 2011, 12(4): 1624-1639.

[2] XIA Y, SHI Z Q. Multi-head attention spatio-temporal convolutional graph network for traffic flow prediction[J]. Application Research of Computers, 2023, 40(3): 776-770 (in Chinese).

[3] REN J H, ZHU Y, MENG X F, et al. Prediction of urban traffic flow using dynamic spatio-temporal neural network [J]. Journal of Chinese Computer Systems, 2023, 44(3): 529-535 (in Chinese).

[4] LI L, HU Z Y, YANG X B. Intelligent analysis of abnormal vehicle behavior based on a digital twin [J]. Journal of Shanghai Jiao Tong University (Science), 2021, 26(5): 587-597.

[5] KIPF T N, WELLING M. Semi-supervised classification with graph convolutional networks [DB/OL]. (2016-09-09). https://arxiv.org/abs/1609.02907

[6] LI Y G, YU R, SHAHABI C, et al. Diffusion convolutional recurrent neural network: Data-driven traffic forecasting [DB/OL]. (2017-07-06). https://arxiv.org/abs/1707.01926

[7] WU Z H, PAN S R, LONG G D, et al. Graph wavenet for deep spatial-temporal graph modeling [C]// 28th International Joint Conference on Artificial Intelligence. Macao: ACM, 2019: 1907-1913.

[8] DRUCKER H, BURGES C J, KAUFMAN L, et al. Duality, geometry, and support vector regression [M]//Advances in neural information processing systems 14. Cambridge: The MIT Press, 2002

[9] MAKRIDAKIS S, HIBON M. ARMA models and the box-jenkins methodology [J]. Journal of Forecasting, 1997, 16(3): 147-163.

[10] YU B, YIN H T, ZHU Z X. Spatio-temporal graph convolutional networks: A deep learning framework for traffic forecasting [C]// Twenty-Seventh International Joint Conference on Artificial Intelligence. Stockholm: IJCAI, 2018: 3634-3640.

[11] BAI L, YAO L N, LI C, et al. Adaptive graph convolutional recurrent network for traffic forecasting [C]// 34th International Conference on Neural Information Processing Systems. Vancouver: NIPS, 2020: 17804-17815.

[12] SONG C, LIN Y F, GUO S N, et al. Spatial-temporal synchronous graph convolutional networks: A new framework for spatial-temporal network data forecasting [J]. Proceedings of the AAAI Conference on Artificial Intelligence, 2020, 34(1): 914-921.

[13] GUO S N, LIN Y F, FENG N, et al. Attention based spatial-temporal graph convolutional networks for traffic flow forecasting [J]. Proceedings of the AAAI Conference on Artificial Intelligence, 2019, 33(1): 922-929.

[14] FANG Z, LONG Q Q, SONG G J, et al. Spatial-temporal graph ODE networks for traffic flow forecasting [C]// 27th ACM SIGKDD Conference on Knowledge Discovery & Data Mining. Singapore. New York: ACM, 2021: 364-373.

[15] WANG Y, ZHU D. SHGCN: A hypergraph-based deep learning model for spatiotemporal traffic flow prediction [C]// 5th ACM SIGSPATIAL International Workshop on AI for Geographic Knowledge Discovery. Seattle: ACM, 2022: 30-39.

[16] LI M Z, ZHU Z X. Spatial-temporal fusion graph neural networks for traffic flow forecasting [J]. Proceedings of the AAAI Conference on Artificial Intelligence, 2021, 35(5): 4189-4196.

[17] LAN S, MA Y, HUANG W, et al. DSTAGNN: Dynamic spatial-temporal aware graph neural network for traffic flow forecasting[C]// 39th International Conference on Machine Learning. San Diego: IMLS, 2022: 11906-11917.

[18] WANG J C, ZHANG Y, WEI Y, et al. Metro passenger flow prediction via dynamic hypergraph convolution networks [J]. IEEE Transactions on Intelligent Transportation Systems, 2021, 22(12): 7891-7903.

[19] FENG Y F, YOU H X, ZHANG Z Z, et al. Hypergraph neural networks [J]. Proceedings of the AAAI Conference on Artificial Intelligence, 2019, 33(1): 3558-3565.

[20] JIANG J W, WEI Y X, FENG Y F, et al. Dynamic hypergraph neural networks [C]// Twenty-Eighth International Joint Conference on Artificial Intelligence. Macao: IJCAI, 2019: 2635-2641.

[21] DING K Z, WANG J L, LI J D, et al. Be more with less: Hypergraph attention networks for inductive text classification [C]// 2020 Conference on Empirical Methods in Natural Language Processing. Stroudsburg: ACL, 2020: 4927-4936.

[22] YADATI N, NIMISHAKAVI M, YADAV P, et al. HyperGCN: A new method of training graph convolutional networks on hypergraphs [DB/OL]. (2018-09-07). https://arxiv.org/abs/1809.02589

[23] YIN N, FENG F L, LUO Z G, et al. Dynamic hypergraph convolutional network [C]//2022 IEEE 38th International Conference on Data Engineering. Kuala Lumpur: IEEE, 2022: 1621-1634.

[24] Wang H, Peng J, Huang F, et al. MICN: Multi-scale local global context modeling for long-term series forecasting[C]// The Eleventh International Conference on Learning Representations. Kigali: ICLR, 2022.

[25] Yin H, Zhang F, Li T R. Short-time traffic flow forecasting based on multi-adjacent graph and multi-head attention mechanism[J]. Computer Science, 2023, 50(4): 40-46 (in Chinese).