Graph-Transformer Based Learning Method for Spatial Structure Representation of Large-scale Urban Road Networks

BAI Fan; HE Wei; HE Kejia; HE Bing

doi:10.11907/rjdk.242065

您当前的位置：

首页 >

文章列表页 >

Graph-Transformer Based Learning Method for Spatial Structure Representation of Large-scale Urban Road Networks

Artificial Intelligence | 更新时间：2026-04-30

- Graph-Transformer Based Learning Method for Spatial Structure Representation of Large-scale Urban Road Networks
- Software Guide Vol. 25, Issue 4, Pages: 65-75(2026)
- 作者机构：
  
  成都信息工程大学计算机学院，四川成都 610225
- 作者简介：
- 基金信息：
- DOI：10.11907/rjdk.242065
  CLC： TP183
- Received：04 March 2025，
  
  Published：15 April 2026
- 稿件说明：
移动端阅览
白帆,何炜,何可佳等.基于Graph-Transformer的大规模城市路网空间结构表示学习方法[J].软件导刊,2026,25(04):65-75.

BAI Fan,HE Wei,HE Kejia,et al.Graph-Transformer Based Learning Method for Spatial Structure Representation of Large-scale Urban Road Networks[J].Software Guide,2026,25(04):65-75.
白帆,何炜,何可佳等.基于Graph-Transformer的大规模城市路网空间结构表示学习方法[J].软件导刊,2026,25(04):65-75. DOI： 10.11907/rjdk.242065.

BAI Fan,HE Wei,HE Kejia,et al.Graph-Transformer Based Learning Method for Spatial Structure Representation of Large-scale Urban Road Networks[J].Software Guide,2026,25(04):65-75. DOI： 10.11907/rjdk.242065.

摘要

路网结构对交通流动效率和安全性具有显著影响。然而，现有路网数据虽然拓扑结构较为完整，但辅助信息极为稀疏，通常难以提取道路网络结构分析所需的特征。因此，如何有效地通过深度学习挖掘结构中隐含的信息成为交通网络分析重要任务之一。提出一种基于Graph-Transformer的道路网络自监督学习模型，模型架构由子图采样数据增强、Graph-Transformer编码器和节点—子图对比损失三部分组成。其中，子图采样旨在提升大规模城市路网分析效率，自监督对比学习则是用于最大化不同子图视图中正负样本对的互信息，以捕获和表示道路网络的隐含结构信息，从而支持下游的道路网络分析任务。实验结果表明，模型在多个城市数据集上的准确率提升0.02%~2.45%，F1分数提升0.07%~2.55%。同时，在下游任务社区划分效果上表现出更优的模块度Q值，提升了0.02%~1.75%。通过研究改进提高了道路网络结构特征提取的有效性，为分析和理解城市大规模路网隐含结构特征提供了一种新方法。

Abstract

The road network structure significantly impacts traffic flow efficiency and safety. However， while existing road network datasets maintain relatively complete topological structures， their auxiliary information is extremely sparse， making it difficult to extract the features required for road network structure analysis.Therefore， effectively mining the implicit information in the structure through deep learning has become one of the important tasks in traffic network analysis. This paper proposes a self-supervised learning model for road networks based on Graph-Transformer. The model architecture consists of three parts： subgraph sampling for data augmentation， a Graph-Transformer encoder， and a node-subgraph contrastive loss. Subgraph sampling is used to improve the efficiency of large-scale urban road network analysis， while self-supervised contrastive learning is employed to maximize the mutual information between positive and negative sample pairs in different subgraph views， thereby capturing and representing the implicit structural information of the road network to support downstream road network analysis tasks. Experimental results demonstrate that the proposed model achieves accuracy improvements of 0.02% to 2.45% and F1 score improvements of 0.07% to 2.55% across multiple city datasets. Additionally， it exhibits better modularity Q values in downstream task community detection and improves by 0.02% to 1.75%. This study enhances the effectiveness of road network structural feature extraction and provides a new method for analyzing and understanding the implicit structural features of large-scale urban road networks.

关键词

Keywords

references

CHEN B F ， WU Z F ， HU W P . Complexity analysis of urban traffic road network based on fractal theory and GIS—taking Guangzhou urban area as an example ［J］. Tropical Geography ， 2011 ， 31 （ 1 ）： 46 - 51 .

陈斌锋，吴志峰，胡伟平 . 基于分形理论与GIS的城区交通路网复杂性分析——以广州市城区为例［J］. 热带地理， 2011 ， 31 （ 1 ）： 46 - 51 .

ZHANG J ， CHEN L . Reliability analysis of road network structure ［J］. Journal of Chang'an University （Natural Science Edition）， 2010 ， 30 （ 4 ）： 77 - 81 .

张菁，陈荔 . 道路网结构可靠性分析［J］. 长安大学学报（自然科学版）， 2010 ， 30 （ 4 ）： 77 - 81 .

PORTA S ， CRUCITTI P ， LATORA V . The network analysis of urban streets： a primal approach ［J］. Environment and Planning B： Planning and Design ， 2006 ， 33 （ 5 ）： 705 - 725 . doi: 10.1068/b32045 http://dx.doi.org/10.1068/b32045

JIANG B . A topological pattern of urban street networks： universality and peculiarity ［J］. Physica A： Statistical Mechanics and its Applications ， 2007 ， 384 （ 2 ）： 647 - 655 . doi: 10.1016/j.physa.2007.05.064 http://dx.doi.org/10.1016/j.physa.2007.05.064

DUAN Y ， LU F . Robustness of city road networks at different granularities ［J］. Physica A： Statistical Mechanics and its Applications ， 2014 ， 411 ： 21 - 34 . doi: 10.1016/j.physa.2014.05.073 http://dx.doi.org/10.1016/j.physa.2014.05.073

JIANG B ， DUAN Y ， LU F ， et al . Topological structure of urban street networks from the perspective of degree correlations ［J］. Environment and Planning B： Planning and Design ， 2014 ， 41 （ 5 ）： 813 - 828 . doi: 10.1068/b39110 http://dx.doi.org/10.1068/b39110

CASALI Y ， HEINIMANN H R . A topological analysis of growth in the Zurich road network ［J］. Computers， Environment and Urban Systems ， 2019 ， 75 ： 244 - 253 . doi: 10.1016/j.compenvurbsys.2019.01.010 http://dx.doi.org/10.1016/j.compenvurbsys.2019.01.010

JIANG B ， LIU C . Street-based topological representations and analyses for predicting traffic flow in GIS ［J］. International Journal of Geographical Information Science ， 2009 ， 23 （ 9 ）： 1119 - 1137 . doi: 10.1080/13658810701690448 http://dx.doi.org/10.1080/13658810701690448

MA D ， OMER I ， OSARAGI T ， et al . Why topology matters in predicting human activities ［J］. Environment and Planning B： Urban Analytics and City Science ， 2019 ， 46 （ 7 ）： 1297 - 1313 . doi: 10.1177/2399808318792268 http://dx.doi.org/10.1177/2399808318792268

GAO S ， WANG Y ， GAO Y ， et al . Understanding urban traffic-flow characteristics： a rethinking of betweenness centrality ［J］. Environment and Planning B： Planning and Design ， 2013 ， 40 （ 1 ）： 135 - 153 . doi: 10.1068/b38141 http://dx.doi.org/10.1068/b38141

JENELIUS E . Network structure and travel patterns： explaining the geographical disparities of road network vulnerability ［J］. Journal of Transport Geography ， 2009 ， 17 （ 3 ）： 234 - 244 . doi: 10.1016/j.jtrangeo.2008.06.002 http://dx.doi.org/10.1016/j.jtrangeo.2008.06.002

ORELLANA D ， GUERRERO M L . Exploring the influence of road network structure on the spatial behaviour of cyclists using crowdsourced data ［J］. Environment and Planning B： Urban Analytics and City Science ， 2019 ， 46 （ 7 ）： 1314 - 1330 . doi: 10.1177/2399808319863810 http://dx.doi.org/10.1177/2399808319863810

ZHUANG D ， WANG Q ， ZHENG Y ， et al . Advancing transportation mode share analysis with built environment： deep hybrid models with urban road network［DB/OL］ . https：//arxiv.org/abs/2405.14079 https://arxiv.org/abs/2405.14079 . doi: 10.2139/ssrn.4764727 http://dx.doi.org/10.2139/ssrn.4764727

ZHENG J ， NI L M . Time-dependent trajectory regression on road networks via multi-task learning ［C］// Proceedings of the AAAI Conference on Artificial Intelligence ， 2013 ： 1048 - 1055 . doi: 10.1609/aaai.v27i1.8577 http://dx.doi.org/10.1609/aaai.v27i1.8577

FRUENSGAARD M ， JEPSEN T S . Improving cost estimation models with estimation updates and road2vec： a feature learning framework for road networks ［EB/OL］. https：//api.semanticscholar.org/CorpusID：44034092 https://api.semanticscholar.org/CorpusID:44034092 .

LIU K ， GAO S ， QIU P ， et al . Road2vec： measuring traffic interactions in urban road system from massive travel routes ［J］. ISPRS International Journal of Geo-Information ， 2017 ， 6 （ 11 ）： 321 . doi: 10.3390/ijgi6110321 http://dx.doi.org/10.3390/ijgi6110321

SHAHABI C ， KOLAHDOUZAN M R ， SHARIFZADEH M . A road network embedding technique for k-nearest neighbor search in moving object databases ［C］// Proceedings of the 10th ACM International Symposium on Advances in Geographic Information Systems ， 2002 ： 94 - 100 . doi: 10.1145/585147.585167 http://dx.doi.org/10.1145/585147.585167

LIU F ， HO Y H ， HUA K A . Privacy protected query processing with road network embedding ［C］// 2011 IEEE International Conference on Advanced Information Networking and Applications ， 2011 ： 481 - 487 . doi: 10.1109/AINA.2011.24 http://dx.doi.org/10.1109/AINA.2011.24

JEPSEN T S ， JENSEN C S ， NIELSEN T D ， et al . On network embedding for machine learning on road networks： a case study on the danish road network ［C］// 2018 IEEE International Conference on Big Data ， 2018 ： 3422 - 3431 . doi: 10.1109/bigdata.2018.8622416 http://dx.doi.org/10.1109/bigdata.2018.8622416

PEROZZI B ， AL-RFOU R ， SKIENA S . Deepwalk： online learning of social representations ［C］// Proceedings of the 20th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining ， 2014 ： 701 - 710 . doi: 10.1145/2623330.2623732 http://dx.doi.org/10.1145/2623330.2623732

GROVER A ， LESKOVEC J . Node2vec： scalable feature learning for networks ［C］// Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining ， 2016 ： 855 - 864 . doi: 10.1145/2939672.2939754 http://dx.doi.org/10.1145/2939672.2939754

TANG J ， QU M ， WANG M ， et al . Line： large-scale information network embedding ［C］// Proceedings of the 24th International Conference on World Wide Web ， 2015 ： 1067 - 1077 . doi: 10.1145/2736277.2741093 http://dx.doi.org/10.1145/2736277.2741093

WANG D ， CUI P ， ZHU W . Structural deep network embedding ［C］// Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining ， 2016 ： 1225 - 1234 . doi: 10.1145/2939672.2939753 http://dx.doi.org/10.1145/2939672.2939753

GOYAL P ， FERRARA E . Graph embedding techniques， applications， and performance： a survey ［J］. Knowledge-Based Systems ， 2018 ， 151 ： 78 - 94 . doi: 10.1016/j.knosys.2018.03.022 http://dx.doi.org/10.1016/j.knosys.2018.03.022

XIE Y ， GONG M ， WANG S ， et al . Sim2vec： node similarity preserving network embedding ［J］. Information Sciences ， 2019 ， 495 ： 37 - 51 . doi: 10.1016/j.ins.2019.05.001 http://dx.doi.org/10.1016/j.ins.2019.05.001

WANG M ， LEE W C ， FU T ， et al . Learning embeddings of intersections on road networks ［C］// Proceedings of the 27th ACM SIGSPATIAL International Conference on Advances in Geographic Information Systems ， 2019 ： 309 - 318 . doi: 10.1145/3347146.3359075 http://dx.doi.org/10.1145/3347146.3359075

HUANG S ， SHAO C ， LI J ， et al . Feature extraction and representation of urban road networks based on travel routes ［J］. Sustainability ， 2020 ， 12 （ 22 ）： 9621 . doi: 10.3390/su12229621 http://dx.doi.org/10.3390/su12229621

JEPSEN T S ， JENSEN C S ， NIELSEN T D . Graph convolutional networks for road networks ［C］// Proceedings of the 27th ACM SIGSPATIAL International Conference on Advances in Geographic Information Systems ， 2019 ： 460 - 463 . doi: 10.1145/3347146.3359094 http://dx.doi.org/10.1145/3347146.3359094

GHARAEE Z ， KOWSHIK S ， STROMANN O ， et al . Graph representation learning for road type classification ［J］. Pattern Recognition ， 2021 ， 120 ： 108174 . doi: 10.1016/j.patcog.2021.108174 http://dx.doi.org/10.1016/j.patcog.2021.108174

NIPPANI A ， LI D ， JU H ， et al . Graph neural networks for road safety modeling： datasets and evaluations for accident analysis［DB/OL］ . https：//arxiv.org/abs/2311.00164 https://arxiv.org/abs/2311.00164 . doi: 10.52202/075280-2265 http://dx.doi.org/10.52202/075280-2265

DWIVEDI V P ， BRESSON X . A generalization of transformer networks to graphs［DB/OL］ . https：//arxiv.org/abs/2012.09699 https://arxiv.org/abs/2012.09699 .

LI C ， XIA L ， REN X ， et al . Graph transformer for recommendation ［C］// Proceedings of the 46th International ACM SIGIR Conference on Research and Development in Information Retrieval ， 2023 ： 1680 - 1689 . doi: 10.1145/3539618.3591723 http://dx.doi.org/10.1145/3539618.3591723

WU Q ， ZHAO W ， LI Z ， et al . Nodeformer： a scalable graph structure learning transformer for node classification ［J］. Advances in Neural Information Processing Systems ， 2022 ， 35 ： 27387 - 27401 . doi: 10.52202/068431-1986 http://dx.doi.org/10.52202/068431-1986

LIU K ， ZHANG M ， XI G ， et al . Enhancing fine-grained intra-urban dengue forecasting by integrating spatial interactions of human movements between urban regions ［J］. PLoS Neglected Tropical Diseases ， 2020 ， 14 （ 12 ）： e0008924 . doi: 10.1371/journal.pntd.0008924 http://dx.doi.org/10.1371/journal.pntd.0008924

SHEN Y ， JIN C ， HUA J ， et al . TTPNet： a neural network for travel time prediction based on tensor decomposition and graph embedding ［J］. IEEE Transactions on Knowledge and Data Engineering ， 2020 ， 34 （ 9 ）： 4514 - 4526 .

CHEN Y ， LI X ， CONG G ， et al . Robust road network representation learning： when traffic patterns meet traveling semantics ［C］// Proceedings of the 30th ACM International Conference on Information & Knowledge Management ， 2021 ： 211 - 220 . doi: 10.1145/3459637.3482293 http://dx.doi.org/10.1145/3459637.3482293

BRIN S ， PAGE L . The anatomy of a large-scale hypertextual web search engine ［J］. Computer Networks and ISDN Systems ， 1998 ， 30 （ 1-7 ）： 107 - 117 . doi: 10.1016/s0169-7552(98)00110-x http://dx.doi.org/10.1016/s0169-7552(98)00110-x

JIAO Y ， XIONG Y ， ZHANG J ， et al . Sub-graph contrast for scalable self-supervised graph representation learning ［C］// 2020 IEEE International Conference on Data Mining ， 2020 ： 222 - 231 . doi: 10.1109/icdm50108.2020.00031 http://dx.doi.org/10.1109/icdm50108.2020.00031

NEWMAN M E J . Modularity and community structure in networks ［J］. Proceedings of the National Academy of Sciences ， 2006 ， 103 （ 23 ）： 8577 - 8582 . doi: 10.1073/pnas.0601602103 http://dx.doi.org/10.1073/pnas.0601602103

BOEING G . Global urban street network data .［EB/OL］. https：//doi.org/10.7910/DVN/3BZ3ZZ https://doi.org/10.7910/DVN/3BZ3ZZ .

KOU S H ， YAO Y ， ZHENG H ， et al . Evaluation of urban traffic layout in China based on road network data and complex graph theory ［J］. Journal of Geo-information Science ， 2021 ， 23 （ 5 ）： 812 - 824 . doi: 10.12082/dqxxkx.2021.200340 http://dx.doi.org/10.12082/dqxxkx.2021.200340

寇世浩，姚尧，郑泓，等 . 基于路网数据和复杂图论的中国城市交通布局评价［J］. 地球信息科学学报， 2021 ， 23 （ 5 ）： 812 - 824 . doi: 10.12082/dqxxkx.2021.200340 http://dx.doi.org/10.12082/dqxxkx.2021.200340

THAKOOR S ， TALLEC C ， AZAR M G ， et al . Bootstrapped representation learning on graphs ［C］// ICLR 2021 Workshop on Geometrical and Topological Representation Learning ， 2021 .

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

HAMILTON W ， YING Z ， LESKOVEC J . Inductive representation learning on large graphs ［C］// Proceedings of the 31st International Conference on Neural Information Processing Systems ， 2017 ： 1025 - 1035 .

Views

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Remaining Time Prediction Methods Oriented to the Educational Process

Adaptive Learning Path Recommendation Supported by Deep Learning——Knowledge Tracing Model Integrating Graph Embedding and Attention Mechanism

A Community Detection Optimization Algorithm Based on GA and SA

GCN-based Android Malware Detection Model

Related Author

LI Ting

XU Xingrong

GAO Tengteng

CHEN Bin

CHU Xu

SUN Xiaoqi

YUAN Yuan

WANG Yan

Related Institution

Zibo Normal University

Zibo Radio and Television Station

School of Educational Science， Nantong University

College of Science， Jiangxi University of Science and Technology

Command & Control Engineering College， Army Engineering University of PLA

⁰