Graph Convolutional Networks for Road Networks

Tobias Skovgaard Jepsen; Christian S. Jensen; Thomas Dyhre Nielsen

doi:10.1145/3347146.3359094

Graph Convolutional Networks for Road Networks

Tobias Skovgaard Jepsen^*, Christian S. Jensen, Thomas Dyhre Nielsen

^*Corresponding author for this work

Research output: Contribution to book/anthology/report/conference proceeding › Article in proceeding › Research › peer-review

27 Citations (Scopus)

380 Downloads (Pure)

Abstract

The application of machine learning techniques in the setting of road networks holds the potential to facilitate many important transportation applications. Graph Convolutional Networks (GCNs) are neural networks that are capable of leveraging the structure of a network. However, many implicit assumptions of GCNs do not apply to road networks.
We introduce the Relational Fusion Network (RFN), a novel type of GCN designed specifically for road networks. In particular, we pro- pose methods that substantially outperform state-of-the-art GCNs on two machine learning tasks in road networks. Furthermore, we show that state-of-the-art GCNs fail to effectively leverage road network structure on these tasks.

Original language	English
Title of host publication	Proceedings of the 27th ACM SIGSPATIAL International Conference on Advances in Geographic Information Systems
Editors	Farnoush Banaei-Kashani, Goce Trajcevski, Ralf Hartmut Guting, Lars Kulik, Shawn Newsam
Number of pages	4
Publisher	Association for Computing Machinery
Publication date	5 Nov 2019
Pages	460-463
ISBN (Print)	9781450369091
ISBN (Electronic)	9781450369091
DOIs	https://doi.org/10.1145/3347146.3359094
Publication status	Published - 5 Nov 2019
Event	27th ACM SIGSPATIAL International Conference on Advances in Geographic Information Systems - 350 W Mart Center Dr, Chicago, United States Duration: 5 Nov 2019 → 8 Nov 2019 https://sigspatial2019.sigspatial.org/

Conference

Conference	27th ACM SIGSPATIAL International Conference on Advances in Geographic Information Systems
Location	350 W Mart Center Dr
Country/Territory	United States
City	Chicago
Period	05/11/2019 → 08/11/2019
Internet address	https://sigspatial2019.sigspatial.org/

Keywords

Road Network
Machine Learning
Graph Representation Learning
Graph Convolutional Networks
Transportation

Access to Document

10.1145/3347146.3359094

PreprintAccepted author manuscript, 799 KBLicence: Unspecified

AUB Link

Search for the material in Aalborg University Library's search engine

27 Citations
1 Article in proceeding
1 Journal article
1 PhD thesis

Relational Fusion Networks: Graph Convolutional Networks for Road Networks
Skovgaard Jepsen, T., Jensen, C. S. & Nielsen, T. D., Jan 2022, In: IEEE Transactions on Intelligent Transportation Systems. 23, 1, p. 418-429 12 p., 9167450.
Research output: Contribution to journal › Journal article › Research › peer-review

Open Access
File
22 Citations (Scopus)

123 Downloads (Pure)
Towards Data-Efficient Mobility Analytics in Spatial Networks
Skovgaard Jepsen, T., 2021, Aalborg Universitetsforlag. 169 p.
Research output: PhD thesis

Open Access
File

78 Downloads (Pure)
Scalable Unsupervised Multi-Criteria Trajectory Segmentation and Driving Preference Mining
Barth, F., Funke, S., Skovgaard Jepsen, T. & Proissl, C., 3 Nov 2020, BIGSPATIAL '20: Proceedings of the 9th ACM SIGSPATIAL International Workshop on Analytics for Big Geospatial Data. Chandola, V., Vatsavai, R. R. & Shashidharan, A. (eds.). Association for Computing Machinery, p. 1-10 10 p. 6
Research output: Contribution to book/anthology/report/conference proceeding › Article in proceeding › Research › peer-review

File
4 Citations (Scopus)

30 Downloads (Pure)

Cite this

Skovgaard Jepsen, T., Jensen, C. S., & Nielsen, T. D. (2019). Graph Convolutional Networks for Road Networks. In F. Banaei-Kashani, G. Trajcevski, R. H. Guting, L. Kulik, & S. Newsam (Eds.), Proceedings of the 27th ACM SIGSPATIAL International Conference on Advances in Geographic Information Systems (pp. 460-463). Association for Computing Machinery. https://doi.org/10.1145/3347146.3359094

Skovgaard Jepsen, Tobias ; Jensen, Christian S. ; Nielsen, Thomas Dyhre. / Graph Convolutional Networks for Road Networks. Proceedings of the 27th ACM SIGSPATIAL International Conference on Advances in Geographic Information Systems. editor / Farnoush Banaei-Kashani ; Goce Trajcevski ; Ralf Hartmut Guting ; Lars Kulik ; Shawn Newsam. Association for Computing Machinery, 2019. pp. 460-463

@inproceedings{e8a8f3050a8546e9904fea23d72159f5,

title = "Graph Convolutional Networks for Road Networks",

abstract = "The application of machine learning techniques in the setting of road networks holds the potential to facilitate many important transportation applications. Graph Convolutional Networks (GCNs) are neural networks that are capable of leveraging the structure of a network. However, many implicit assumptions of GCNs do not apply to road networks.We introduce the Relational Fusion Network (RFN), a novel type of GCN designed specifically for road networks. In particular, we pro- pose methods that substantially outperform state-of-the-art GCNs on two machine learning tasks in road networks. Furthermore, we show that state-of-the-art GCNs fail to effectively leverage road network structure on these tasks.",

keywords = "Road Network, Machine Learning, Graph Representation Learning, Graph Convolutional Networks, Transportation",

author = "{Skovgaard Jepsen}, Tobias and Jensen, {Christian S.} and Nielsen, {Thomas Dyhre}",

year = "2019",

month = nov,

day = "5",

doi = "10.1145/3347146.3359094",

language = "English",

isbn = "9781450369091",

pages = "460--463",

editor = "Farnoush Banaei-Kashani and Goce Trajcevski and Guting, {Ralf Hartmut} and Lars Kulik and Shawn Newsam",

booktitle = "Proceedings of the 27th ACM SIGSPATIAL International Conference on Advances in Geographic Information Systems",

publisher = "Association for Computing Machinery",

address = "United States",

note = "27th ACM SIGSPATIAL International Conference on Advances in Geographic Information Systems, ACM SIGSPATIAL 2019 ; Conference date: 05-11-2019 Through 08-11-2019",

url = "https://sigspatial2019.sigspatial.org/",

}

Skovgaard Jepsen, T, Jensen, CS & Nielsen, TD 2019, Graph Convolutional Networks for Road Networks. in F Banaei-Kashani, G Trajcevski, RH Guting, L Kulik & S Newsam (eds), Proceedings of the 27th ACM SIGSPATIAL International Conference on Advances in Geographic Information Systems. Association for Computing Machinery, pp. 460-463, 27th ACM SIGSPATIAL International Conference on Advances in Geographic Information Systems, Chicago, Illinois, United States, 05/11/2019. https://doi.org/10.1145/3347146.3359094

Graph Convolutional Networks for Road Networks. / Skovgaard Jepsen, Tobias; Jensen, Christian S.; Nielsen, Thomas Dyhre.
Proceedings of the 27th ACM SIGSPATIAL International Conference on Advances in Geographic Information Systems. ed. / Farnoush Banaei-Kashani; Goce Trajcevski; Ralf Hartmut Guting; Lars Kulik; Shawn Newsam. Association for Computing Machinery, 2019. p. 460-463.

Research output: Contribution to book/anthology/report/conference proceeding › Article in proceeding › Research › peer-review

TY - GEN

T1 - Graph Convolutional Networks for Road Networks

AU - Skovgaard Jepsen, Tobias

AU - Jensen, Christian S.

AU - Nielsen, Thomas Dyhre

PY - 2019/11/5

Y1 - 2019/11/5

N2 - The application of machine learning techniques in the setting of road networks holds the potential to facilitate many important transportation applications. Graph Convolutional Networks (GCNs) are neural networks that are capable of leveraging the structure of a network. However, many implicit assumptions of GCNs do not apply to road networks.We introduce the Relational Fusion Network (RFN), a novel type of GCN designed specifically for road networks. In particular, we pro- pose methods that substantially outperform state-of-the-art GCNs on two machine learning tasks in road networks. Furthermore, we show that state-of-the-art GCNs fail to effectively leverage road network structure on these tasks.

AB - The application of machine learning techniques in the setting of road networks holds the potential to facilitate many important transportation applications. Graph Convolutional Networks (GCNs) are neural networks that are capable of leveraging the structure of a network. However, many implicit assumptions of GCNs do not apply to road networks.We introduce the Relational Fusion Network (RFN), a novel type of GCN designed specifically for road networks. In particular, we pro- pose methods that substantially outperform state-of-the-art GCNs on two machine learning tasks in road networks. Furthermore, we show that state-of-the-art GCNs fail to effectively leverage road network structure on these tasks.

KW - Road Network

KW - Machine Learning

KW - Graph Representation Learning

KW - Graph Convolutional Networks

KW - Transportation

U2 - 10.1145/3347146.3359094

DO - 10.1145/3347146.3359094

M3 - Article in proceeding

SN - 9781450369091

SP - 460

EP - 463

BT - Proceedings of the 27th ACM SIGSPATIAL International Conference on Advances in Geographic Information Systems

A2 - Banaei-Kashani, Farnoush

A2 - Trajcevski, Goce

A2 - Guting, Ralf Hartmut

A2 - Kulik, Lars

A2 - Newsam, Shawn

PB - Association for Computing Machinery

T2 - 27th ACM SIGSPATIAL International Conference on Advances in Geographic Information Systems

Y2 - 5 November 2019 through 8 November 2019

ER -

Graph Convolutional Networks for Road Networks

Abstract

Conference

Keywords

Access to Document

AUB Link

Fingerprint

Research output

Relational Fusion Networks: Graph Convolutional Networks for Road Networks

Towards Data-Efficient Mobility Analytics in Spatial Networks

Scalable Unsupervised Multi-Criteria Trajectory Segmentation and Driving Preference Mining

Cite this