Multi-graph diffusion attention network for traffic flow prediction

doi:10.11772/j.issn.1001-9081.2024050636

Journal of Computer Applications ›› 2025, Vol. 45 ›› Issue (5): 1472-1479.DOI: 10.11772/j.issn.1001-9081.2024050636

• Artificial intelligence • Previous Articles

Multi-graph diffusion attention network for traffic flow prediction

Quan WANG¹^,²(), Qixiang LU¹, Pei SHI²

^1.School of Electronics and Information Engineering，Nanjing University of Information Science and Technology，Nanjing Jiangsu 210044，China
^2.School of IoT Engineering，Wuxi University，Wuxi Jiangsu 214105，China

Received:2024-05-17 Revised:2024-08-01 Accepted:2024-08-01 Online:2024-08-20 Published:2025-05-10
Contact: Quan WANG
About author:WANG Quan， born in 1980， Ph. D.， professor of engineering. His research interests include internet of vehicles， industrial internet of things.
LU Qixiang， born in 2000， M. S. candidate. His research interests include internet of vehicles， traffic flow prediction.
SHI Pei， born in 1988， Ph. D.， lecturer. Her research interests include wireless sensor network， agricultural internet of things.
Supported by:
National Natural Science Foundation of China(62072216);Jiangsu Province Natural Science Project of Higher Educational Institutions(21KJB520020);Wuxi Municipal Science and Technology “Taihu Light” Basic Research Program(K20231021)

用于交通流量预测的多图扩散注意力网络

王泉¹^,²(), 陆啟想¹, 施珮²

^1.南京信息工程大学电子与信息工程学院，南京 210044
^2.无锡学院物联网工程学院，江苏无锡 214105

通讯作者: 王泉
作者简介:王泉（1980—），男，湖北咸宁人，正高级工程师，博士，主要研究方向：车联网、工业物联网
陆啟想（2000—），男，江苏宿迁人，硕士研究生，主要研究方向：车联网、交通流量预测
施珮（1988—），女，安徽宣城人，讲师，博士，主要研究方向：无线传感网络、农业物联网。
基金资助:
国家自然科学基金资助项目(62072216);江苏省高校自然科学研究面上项目(21KJB520020);无锡市“太湖之光”科技攻关计划基础研究项目(K20231021)

Abstract

Abstract:

Current traffic flow prediction methods based on spatio-temporal feature extraction has problems of insufficient capture of global spatial correlation and dynamic long-term temporal dependency， where spatial correlation mining relies on the quality of graph structure heavily. Therefore， a Multi-Graph Diffusion Attention Network （MGDAN） was proposed， consisting of a Multi-Graph Diffusion Attention Module （MGDAM） and a temporal attention module. Firstly， adaptive spatio-temporal embedding generator was used to construct dynamic spatio-temporal information. Secondly， a Maximal Information Coefficient （MIC） matrix and an adaptive matrix were utilized to explore fine-grained spatial information， and a global spatial attention mechanism was employed to capture dynamic spatial correlation. Finally， the temporal attention module was used to extract nonlinear temporal correlation， and the integration of the three modules was carried out to realize effective extraction of spatio-temporal correlation. Experimental results demonstrate that， on PEMS08 dataset， the Mean Absolute Error （MAE） of MGDAN model within one hour has 19.34% and 5.74% reductions compared to those of Spatio-Temporal AutoEncoder （ST_AE） and Spatial-Temporal IDentity （STID） models， respectively. At the same time， MGDAN model outperforms 9 baseline models in overall prediction performance， and can conduct medium- and long-term traffic flow prediction accurately， providing theoretical basis for urban traffic dispersion.

Key words: traffic flow prediction, spatio-temporal model, adaptive spatio-temporal embedding, Graph Convolutional Network (GCN), attention network

摘要：

当前基于时空特征提取的交通流量预测方法中存在挖掘全局空间相关性与长期的动态时间依赖关系能力不足的问题，其中空间相关性的挖掘很大程度上取决于图结构的质量，为此提出一种多图扩散注意力网络（MGDAN），主要包括多图扩散注意力模块（MGDAM）和时间注意力模块。首先，使用自适应时空嵌入生成器构建动态的时空信息；其次，采用最大互信息系数（MIC）矩阵与自适应矩阵挖掘细粒度的空间信息，并利用全局空间注意力机制挖掘动态的空间相关性；最后，使用时间注意力模块提取非线性的时间相关性，并通过3个模块的结合实现时空相关性的有效提取。在PEMS08数据集上的实验结果表明，MGDAN在1 h内的平均绝对误差（MAE）相较于时空自编码器（ST_AE）和时空身份信息（STID）模型分别降低了19.34%和5.74%，且整体预测性能均优于9个基线模型，能够精准地进行中长期交通流量预测，为城市交通疏导提供理论依据。

关键词: 交通流量预测, 时空模型, 自适应时空嵌入, 图卷积网络, 注意力网络

CLC Number:

TP181

Quan WANG, Qixiang LU, Pei SHI. Multi-graph diffusion attention network for traffic flow prediction[J]. Journal of Computer Applications, 2025, 45(5): 1472-1479.

王泉, 陆啟想, 施珮. 用于交通流量预测的多图扩散注意力网络[J]. 《计算机应用》唯一官方网站, 2025, 45(5): 1472-1479.

Figures/Tables 13

Fig. 1 Overall structure of MGDAN

Fig. 2 Adaptive spatio-temporal embedding generator

Fig. 3 Structure of MGDAM

Tab.1 Dataset details

数据集	节点数	步长	采样频率/min	目标特征
PEMS04	307	16 992	5	交通流量
PEMS08	170	17 856	5	交通流量

Tab.2 Average performance comparison of different models on PEMS04 and PEMS08 datasets

数据集	模型	MAE	RMSE	MAPE/%
PEMS04	DCRNN	20.73	33.23	14.61
	STGCN	20.07	31.89	13.83
	Graph WaveNet	19.52	30.87	14.57
	STSGCN	21.65	34.01	14.50
	ASTGCN	21.60	34.34	14.00
	GMAN	19.79	31.58	14.63
	DDSTGCN	19.61	31.05	13.70
	RGDAN	19.25	32.19	13.12
	STID	18.55	30.20	12.16
	ST_AE	19.97	31.48	16.04
	MGDAN	18.30	30.59	12.03
PEMS08	DCRNN	16.56	25.79	12.53
	STGCN	16.07	27.83	11.40
	Graph WaveNet	15.01	23.57	10.00
	STSGCN	18.55	28.43	11.86
	ASTGCN	18.41	28.47	11.00
	GMAN	15.69	24.20	10.99
	DDSTGCN	15.30	24.16	10.53
	RGDAN	15.46	24.31	10.13
	STID	14.29	23.59	9.25
	ST_AE	16.70	25.94	12.04
	MGDAN	13.47	23.13	8.82

Fig. 4 Comparison between predicted and true values of nodes at 1 000 consecutive steps

Fig. 5 Comparison of 12-step prediction results of different models on PEMS04 dataset

Fig. 6 Comparison of 12-step prediction results of different models on PEMS08 dataset

Tab. 3 Results of ablation experiments

数据集	模型	MAE	RMSE	MAPE/%
PEMS04	w/o $E a$	19.45	31.26	13.04
	w/o $A M I C$	18.71	30.82	12.34
	w/o TA	18.98	30.52	12.65
	w/o SA	18.78	30.97	12.15
	MGDAN	18.30	30.59	12.03
PEMS08	w/o $E a$	14.87	25.85	9.71
	w/o $A M I C$	13.97	23.45	9.35
	w/o TA	14.23	24.86	9.58
	w/o SA	14.11	23.41	9.50
	MGDAN	13.47	23.13	8.82

Tab. 3 Results of ablation experiments

数据集	模型	MAE	RMSE	MAPE/%
PEMS04	w/o $E a$	19.45	31.26	13.04
	w/o $A M I C$	18.71	30.82	12.34
	w/o TA	18.98	30.52	12.65
	w/o SA	18.78	30.97	12.15
	MGDAN	18.30	30.59	12.03
PEMS08	w/o $E a$	14.87	25.85	9.71
	w/o $A M I C$	13.97	23.45	9.35
	w/o TA	14.23	24.86	9.58
	w/o SA	14.11	23.41	9.50
	MGDAN	13.47	23.13	8.82

Tab. 4 Influence of attention head number on model performance

数据集	H	MAE	RMSE	MAPE/%
PEMS04	1	18.91	30.52	12.65
	2	18.58	30.36	12.27
	4	18.30	30.59	12.03
	8	18.46	30.72	11.78
	16	18.71	30.29	12.38
PEMS08	1	13.91	23.27	10.01
	2	13.70	23.22	9.20
	4	13.47	23.13	8.82
	8	13.52	23.17	8.88
	16	13.89	23.51	9.14

Fig. 7 Performance under different Dropout and L

Tab. 5 Computing time on PEMS08 dataset

模型	计算时间
模型	每轮训练的时间/s	推理时间/s
DCRNN	121.3	12.5
GMAN	210.6	14.7
DDSTGCN	60.2	4.5
RGDAN	122.3	4.8
MGDAN	86.8	5.8

Fig. 8 Visualization of AMIC and Aadp

References 24

1	LI Y， YU R， SHAHABI C， et al. Diffusion convolutional recurrent neural network： data-driven traffic forecasting［EB/OL］. ［2024-01-19］. .
2	YU B， YIN H， ZHU Z. Spatio-temporal graph convolutional networks： a deep learning framework for traffic forecasting［C］// Proceedings of the 27th International Joint Conference on Artificial Intelligence. San Francisco： Morgan Kaufmann Publishers Inc.， 2018： 3634-3640.
3	SONG C， LIN Y， GUO S， et al. Spatial-temporal synchronous graph convolutional networks： a new framework for spatial-temporal network data forecasting［C］// Proceedings of the 34th AAAI Conference on Artificial Intelligence. Palo Alto： AAAI Press， 2020： 914-921.
4	LI M， ZHU Z. Spatial-temporal fusion graph neural networks for traffic flow forecasting［C］// Proceedings of the 35th AAAI Conference on Artificial Intelligence. Palo Alto： AAAI Press， 2021： 4189-4196.
5	NI Q， ZHANG M. STGMN： a gated multi-graph convolutional network framework for traffic flow prediction［J］. Applied Intelligence， 2022， 52（13）： 15026-15039.
6	YU K， QIN X， JIA Z， et al. Cross-attention fusion based spatial-temporal multi-graph convolutional network for traffic flow prediction［J］. Sensors， 2021， 21（24）： No.8468.
7	BAO Y， HUANG J， SHEN Q， et al. Spatial-temporal complex graph convolution network for traffic flow prediction［J］. Engineering Applications of Artificial Intelligence， 2023， 121： No.106044.
8	HAN L， DU B， SUN L， et al. Dynamic and multi-faceted spatio-temporal deep learning for traffic speed forecasting［C］// Proceedings of the 27th ACM SIGKDD Conference on Knowledge Discovery and Data Mining. New York： ACM， 2021： 547-555.
9	SUN Y， JIANG X， HU Y， et al. Dual dynamic spatial-temporal graph convolution network for traffic prediction［J］. IEEE Transactions on Intelligent Transportation Systems， 2022， 23（12）： 23680-23693.
10	BAI L， YAO L， LI C， et al. Adaptive graph convolutional recurrent network for traffic forecasting［C］// Proceedings of the 34th International Conference on Neural Information Processing Systems. Red Hook： Curran Associates Inc.， 2020： 17804-17815.
11	徐丽，符祥远，李浩然.基于门控卷积的时空交通流预测模型［J］.计算机应用，2023，43（9）：2760-2765.
	XU L， FU X Y， LI H R. Spatio-temporal traffic flow prediction model based on gated convolution［J］. Journal of Computer Applications， 2023， 43（9）： 2760-2765.
12	ZHAO L， SONG Y， ZHANG C， et al. T-GCN： a temporal graph convolutional network for traffic prediction［J］. IEEE Transactions on Intelligent Transportation Systems， 2020， 21（9）： 3848-3858.
13	顾焰杰，张英俊，刘晓倩，等.基于时空多图融合的交通流量预测［J］.计算机应用，2024，44（8）：2618-2625.
	GU Y J， ZHANG Y J， LIU X Q， et al. Traffic flow forecasting via spatio-temporal multi-graph fusion［J］. Journal of Computer Applications， 2024， 44（8）： 2618-2625.
14	CHEN L， SHI P， LI G， et al. Traffic flow prediction using multi-view graph convolution and masked attention mechanism［J］. Computer Communications， 2022， 194： 446-457.
15	WU Z， PAN S， LONG G， et al. Graph WaveNet for deep spatial-temporal graph modeling［C］// Proceedings of the 28th International Joint Conference on Artificial Intelligence. San Francisco： Morgan Kaufmann Publishers Inc.， 2019： 1907-1913.
16	VASWANI A， SHAZEER N， PARMAR N， et al. Attention is all you need［C］// Proceedings of the 31st International Conference on Neural Information Processing Systems. Red Hook： Curran Associates Inc.， 2017： 6000-6010.
17	LIU M， ZHU T， YE J， et al. Spatio-temporal AutoEncoder for traffic flow prediction［J］. IEEE Transactions on Intelligent Transportation Systems， 2023， 24（5）： 5516-5526.
18	SHAO Z， ZHANG Z， WANG F， et al. Spatial-temporal identity： a simple yet effective baseline for multivariate time series forecasting［C］// Proceedings of the 31st ACM International Conference on Information and Knowledge Management. New York： ACM， 2022： 4454-4458.
19	ZHANG H， WANG H， CHEN L， et al. Traffic flow forecasting based on Transformer with diffusion graph attention network［J］. International Journal of Automotive Technology， 2024， 25（3）：455-468.
20	SU J， JIN Z， REN J， et al. GDFormer： a graph diffusing attention based approach for traffic flow prediction［J］. Pattern Recognition Letters， 2022，156： 126-132.
21	ZHENG C， FAN X， WANG C， et al. GMAN： a graph multi-attention network for traffic prediction［C］// Proceedings of the 34th AAAI Conference on Artificial Intelligence. Palo Alto： AAAI Press， 2020： 1234-1241.
22	FAN J， WENG W， TIAN H， et al. RGDAN： a random graph diffusion attention network for traffic prediction［J］. Neural Networks， 2024， 172： No.106093.
23	CHEN C， PETTY K， SKABARDONIS A， et al. Freeway performance measurement system： mining loop detector data［J］. Transportation Research Record， 2001， 1748（1）： 96-102.
24	GUO S， LIN Y， FENG N， et al. Attention based spatial-temporal graph convolutional networks for traffic flow forecasting［C］// Proceedings of the 33rd AAAI Conference on Artificial Intelligence. Palo Alto： AAAI Press， 2019： 922-929.

[1]	Yufei LONG, Yuchen MOU, Ye LIU. Multi-source data representation learning model based on tensorized graph convolutional network and contrastive learning [J]. Journal of Computer Applications, 2025, 45(5): 1372-1378.
[2]	Weichao DANG, Chujun SONG, Gaimei GAO, Chunxia LIU. Multi-behavior recommendation based on cascading residual graph convolutional network [J]. Journal of Computer Applications, 2025, 45(4): 1223-1231.
[3]	Jiaxin LI, Site MO. Power work order classification in substation area based on MiniRBT-LSTM-GAT and label smoothing [J]. Journal of Computer Applications, 2025, 45(4): 1356-1362.
[4]	Kun FU, Shicong YING, Tingting ZHENG, Jiajie QU, Jingyuan CUI, Jianwei LI. Graph data augmentation method for few-shot node classification [J]. Journal of Computer Applications, 2025, 45(2): 392-402.
[5]	Liang ZHU, Jingzhe MU, Hongqiang ZUO, Jingzhong GU, Fubao ZHU. Location privacy-preserving recommendation scheme based on federated graph neural network [J]. Journal of Computer Applications, 2025, 45(1): 136-143.
[6]	Jianpeng HU, Lichen ZHANG. Deep spatio-temporal network model for multi-time step wind power prediction [J]. Journal of Computer Applications, 2025, 45(1): 98-105.
[7]	Xin YANG, Xueni CHEN, Chunjiang WU, Shijie ZHOU. Short-term traffic flow prediction of urban highway based on variant residual model and Transformer [J]. Journal of Computer Applications, 2024, 44(9): 2947-2951.
[8]	Chuanlin PANG, Rui TANG, Ruizhi ZHANG, Chuan LIU, Jia LIU, Shibo YUE. Distributed power allocation algorithm based on graph convolutional network for D2D communication systems [J]. Journal of Computer Applications, 2024, 44(9): 2855-2862.
[9]	Guixiang XUE, Hui WANG, Weifeng ZHOU, Yu LIU, Yan LI. Port traffic flow prediction based on knowledge graph and spatio-temporal diffusion graph convolutional network [J]. Journal of Computer Applications, 2024, 44(9): 2952-2957.
[10]	Hang YANG, Wanggen LI, Gensheng ZHANG, Zhige WANG, Xin KAI. Multi-layer information interactive fusion algorithm based on graph neural network for session-based recommendation [J]. Journal of Computer Applications, 2024, 44(9): 2719-2725.
[11]	Huanhuan LI, Tianqiang HUANG, Xuemei DING, Haifeng LUO, Liqing HUANG. Public traffic demand prediction based on multi-scale spatial-temporal graph convolutional network [J]. Journal of Computer Applications, 2024, 44(7): 2065-2072.
[12]	Tianci KE, Jianhua LIU, Shuihua SUN, Zhixiong ZHENG, Zijie CAI. Aspect-level sentiment analysis model combining strong association dependency and concise syntax [J]. Journal of Computer Applications, 2024, 44(6): 1786-1795.
[13]	Shibin LI, Jun GONG, Shengjun TANG. Semi-supervised heterophilic graph representation learning model based on Graph Transformer [J]. Journal of Computer Applications, 2024, 44(6): 1816-1823.
[14]	Boyue WANG, Yingxiang LI, Jiandan ZHONG. Segmentation network for day and night ground-based cloud images based on improved Res-UNet [J]. Journal of Computer Applications, 2024, 44(4): 1310-1316.
[15]	Longtao GAO, Nana LI. Aspect sentiment triplet extraction based on aspect-aware attention enhancement [J]. Journal of Computer Applications, 2024, 44(4): 1049-1057.

Multi-graph diffusion attention network for traffic flow prediction

用于交通流量预测的多图扩散注意力网络

RichHTML

PDF

Knowledge

Abstract

Cite this article

share this article

Figures/Tables 13

References 24

Related Articles 15

Recommended Articles

Metrics