Full Text:   <1367>

Summary:  <575>

CLC number: TP391

On-line Access: 2021-09-10

Received: 2020-05-21

Revision Accepted: 2021-02-17

Crosschecked: 2021-04-01

Cited: 0

Clicked: 1880

Citations:  Bibtex RefMan EndNote GB/T7714

 ORCID:

Dewen Seng

https://orcid.org/0000-0003-0921-848X

Xiaoying Shi

https://orcid.org/0000-0003-0452-2503

-   Go to

Article info.
Open peer comments

Frontiers of Information Technology & Electronic Engineering  2021 Vol.22 No.9 P.1179-1193

http://doi.org/10.1631/FITEE.2000243


Forecasting traffic flows in irregular regions with multi-graph convolutional network and gated recurrent unit


Author(s):  Dewen Seng, Fanshun Lv, Ziyi Liang, Xiaoying Shi, Qiming Fang

Affiliation(s):  School of Computer Science and Technology, Hangzhou Dianzi University, Hangzhou 310018, China

Corresponding email(s):   sengdw@hdu.edu.cn, 172050041@hdu.edu.cn, liangziyi2020@163.com, shixiaoying@hdu.edu.cn, fangqiming@hdu.edu.cn

Key Words:  Traffic flow prediction, Multi-graph convolutional network, Gated recurrent unit, Irregular regions


Dewen Seng, Fanshun Lv, Ziyi Liang, Xiaoying Shi, Qiming Fang. Forecasting traffic flows in irregular regions with multi-graph convolutional network and gated recurrent unit[J]. Frontiers of Information Technology & Electronic Engineering, 2021, 22(9): 1179-1193.

@article{title="Forecasting traffic flows in irregular regions with multi-graph convolutional network and gated recurrent unit",
author="Dewen Seng, Fanshun Lv, Ziyi Liang, Xiaoying Shi, Qiming Fang",
journal="Frontiers of Information Technology & Electronic Engineering",
volume="22",
number="9",
pages="1179-1193",
year="2021",
publisher="Zhejiang University Press & Springer",
doi="10.1631/FITEE.2000243"
}

%0 Journal Article
%T Forecasting traffic flows in irregular regions with multi-graph convolutional network and gated recurrent unit
%A Dewen Seng
%A Fanshun Lv
%A Ziyi Liang
%A Xiaoying Shi
%A Qiming Fang
%J Frontiers of Information Technology & Electronic Engineering
%V 22
%N 9
%P 1179-1193
%@ 2095-9184
%D 2021
%I Zhejiang University Press & Springer
%DOI 10.1631/FITEE.2000243

TY - JOUR
T1 - Forecasting traffic flows in irregular regions with multi-graph convolutional network and gated recurrent unit
A1 - Dewen Seng
A1 - Fanshun Lv
A1 - Ziyi Liang
A1 - Xiaoying Shi
A1 - Qiming Fang
J0 - Frontiers of Information Technology & Electronic Engineering
VL - 22
IS - 9
SP - 1179
EP - 1193
%@ 2095-9184
Y1 - 2021
PB - Zhejiang University Press & Springer
ER -
DOI - 10.1631/FITEE.2000243


Abstract: 
The prediction of regional traffic flows is important for traffic control and management in an intelligent traffic system. With the help of deep neural networks, the convolutional neural network or residual neural network, which can be applied only to regular grids, is adopted to capture the spatial dependence for flow prediction. However, the obtained regions are always irregular considering the road network and administrative boundaries; thus, dividing the city into grids is inaccurate for prediction. In this paper, we propose a new model based on multi-graph convolutional network and gated recurrent unit (MGCN-GRU) to predict traffic flows for irregular regions. Specifically, we first construct heterogeneous inter-region graphs for a city to reflect the relationships among regions. In each graph, nodes represent the irregular regions and edges represent the relationship types between regions. Then, we propose a multi-graph convolutional network to fuse different inter-region graphs and additional attributes. The GRU is further used to capture the temporal dependence and to predict future traffic flows. Experimental results based on three real-world large-scale datasets (public bicycle system dataset, taxi dataset, and dockless bike-sharing dataset) show that our MGCN-GRU model outperforms a variety of existing methods.

基于多图卷积网络和门控循环单元的不规则区域交通流量预测

僧德文,吕凡顺,梁紫怡,史晓颖,方启明
杭州电子科技大学计算机学院,中国杭州市,310018
摘要:区域交通流量预测对智能交通系统的交通控制和管理十分重要。借助深度神经网络,采用仅适用于规则网格的循环神经网络或残差神经网络捕获流量预测的空间依赖性。但是,考虑到路网和行政边界得到的区域通常是不规则的。因此将城市划分成网格进行预测是不准确的。提出一种基于多图卷积网络和门控循环单元(MGCN-GRU)的不规则区域交通流量预测模型。首先,构建一个城市异质区域间关联图反映各区域间的关联。在每个图中,节点表示不规则区域,边代表区域间的关联类型。然后,提出一个多图卷积网络融合不同区域间关联图和附加属性。进一步采用门控循环单元捕获时序依赖并预测未来交通流量。实验结果表明,基于3个真实大数据集(公共自行车系统数据集、出租车数据集和无桩共享自行车数据集),所提MGCN-GRU模型性能优于多个现有方法。

关键词:交通流量预测;多图卷积网络;门控循环单元;不规则区域

Darkslateblue:Affiliate; Royal Blue:Author; Turquoise:Article

Reference

[1]Box GEP, Jenkins GM, Reinsel GC, 2015. Time Series Analysis: Forecasting and Control. John Wiley & Sons, New York, USA.

[2]Bruna J, Zaremba W, Szlam A, et al., 2014. Spectral networks and locally connected networks on graphs. Proc Int Conf on Learning Representations, p.1-14.

[3]Chai D, Wang LY, Yang Q, 2018. Bike flow prediction with multi-graph convolutional networks. Proc 26th ACM SIGSPATIAL Int Conf on Advances in Geographic Information Systems, p.397-400.

[4]Chandra SR, Al-Deek H, 2009. Predictions of freeway traffic speeds and volumes using vector autoregressive models. J Intell Transp Syst, 13(2):53-72.

[5]Defferrard M, Bresson X, Vandergheynst P, 2016. Convolutional neural networks on graphs with fast localized spectral filtering. Proc 30th Int Conf on Neural Information Processing Systems, p.3844-3852.

[6]Fu R, Zhang Z, Li L, 2016. Using LSTM and GRU neural network methods for traffic flow prediction. Proc 31st Youth Academic Annual Conf of Chinese Association of Automation, p.324-328.

[7]Kaltenbrunner A, Meza R, Grivolla J, et al., 2010. Urban cycles and mobility patterns: exploring and predicting trends in a bicycle-based public transport system. Perv Mob Comput, 6(4):455-466.

[8]Kim Y, Wang P, Mihaylova L, 2019. Scalable learning with a structural recurrent neural network for short-term traffic prediction. IEEE Sens J, 19(23):11359-11366.

[9]Kipf TN, Welling M, 2017. Semi-supervised classification with graph convolutional networks. Proc 5th Int Conf on Learning Representations, p.1-10.

[10]Li YG, Yu R, Shahabi C, et al., 2018. Diffusion convolutional recurrent neural network: data-driven traffic forecasting. Proc 6th Int Conf on Learning Representations, p.1-10.

[11]Monti F, Bronstein MM, Bresson X, 2017. Geometric matrix completion with recurrent multi-graph neural networks. Proc 31st Int Conf on Neural Information Processing Systems, p.3697-3707.

[12]Moreira-Matias L, Gama J, Ferreira M, et al., 2013. Predicting taxi–passenger demand using streaming data. IEEE Trans Intell Transp Syst, 14(3):1393-1402.

[13]Seo Y, Defferrard M, Vandergheynst P, et al., 2018. Structured sequence modeling with graph convolutional recurrent networks. Proc 25th Int Conf on Neural Information, p.362-373.

[14]Tian YX, Pan L, 2015. Predicting short-term traffic flow by long short-term memory recurrent neural network. IEEE Int Conf on Smart City/SocialCom/SustainCom, p.153-158.

[15]Wang P, Kim Y, Vaci L, et al., 2018. Short-term traffic prediction with vicinity Gaussian process in the presence of missing data. Sensor Data Fusion: Trends, Solutions, Applications, p.1-6.

[16]Williams BM, Hoel LA, 2003. Modeling and forecasting vehicular traffic flow as a seasonal ARIMA process: theoretical basis and empirical results. J Transp Eng, 129(6):664-672.

[17]Yao HX, Wu F, Ke JT, et al., 2018. Deep multi-view spatial-temporal network for taxi demand prediction. Proc 32nd AAAI Conf on Artificial Intelligence, p.2588-2595.

[18]Ying R, He RN, Chen KF, et al., 2018. Graph convolutional neural networks for web-scale recommender systems. Proc 24th ACM SIGKDD Int Conf on Knowledge Discovery & Data Mining, p.974-983.

[19]Yoon JW, Pinelli F, Calabrese F, 2012. Cityride: a predictive bike sharing journey advisor. Proc 13th Int Conf on Mobile Data Management, p.306-311.

[20]Yu B, Yin HT, Zhu ZX, 2018. Spatio-temporal graph convolutional networks: a deep learning framework for traffic forecasting. Proc 27th Int Joint Conf on Artificial Intelligence, p.1-7.

[21]Yu R, Li YG, Shahabi C, et al., 2017. Deep learning: a generic approach for extreme condition traffic forecasting. Proc SIAM Int Conf on Data Mining. p.777-785.

[22]Yuan NJ, Zheng Y, Xie X, et al., 2015. Discovering urban functional zones using latent activity trajectories. IEEE Trans Knowl Data Eng, 27(3):712-725.

[23]Zhang JB, Zheng Y, Qi DK, et al., 2016. DNN-based prediction model for spatio-temporal data. Proc 24th ACM SIGSPATIAL Int Conf on Advances in Geographic Information Systems, p.92.

[24]Zhang JB, Zheng Y, Qi DK, et al., 2018. Predicting citywide crowd flows using deep spatio-temporal residual networks. Artif Intell, 259:147-166.

[25]Zhao L, Song YJ, Zhang C, et al., 2020. T-GCN: a temporal graph convolutional network for traffic prediction. IEEE Trans Intell Transp Syst, 21(9):3848-3858.

[26]Zhu L, Yu FR, Wang YG, et al., 2019. Big data analytics in intelligent transportation systems: a survey. IEEE Trans Intell Transp Syst, 20(1):383-398.

Open peer comments: Debate/Discuss/Question/Opinion

<1>

Please provide your name, email address and a comment





Journal of Zhejiang University-SCIENCE, 38 Zheda Road, Hangzhou 310027, China
Tel: +86-571-87952783; E-mail: cjzhang@zju.edu.cn
Copyright © 2000 - Journal of Zhejiang University-SCIENCE