Graph Convolutional Networks for Road Networks

Research output: Contribution to book/anthology/report/conference proceedingArticle in proceedingResearchpeer-review

2 Citations (Scopus)
152 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 languageEnglish
Title of host publicationProceedings of the 27th ACM SIGSPATIAL International Conference on Advances in Geographic Information Systems
EditorsFarnoush Banaei-Kashani, Goce Trajcevski, Ralf Hartmut Guting, Lars Kulik, Shawn Newsam
Number of pages4
PublisherAssociation for Computing Machinery
Publication date5 Nov 2019
Pages460-463
ISBN (Print)9781450369091
ISBN (Electronic)9781450369091
DOIs
Publication statusPublished - 5 Nov 2019
Event27th ACM SIGSPATIAL International Conference on Advances in Geographic Information Systems - 350 W Mart Center Dr, Chicago, United States
Duration: 5 Nov 20198 Nov 2019
https://sigspatial2019.sigspatial.org/

Conference

Conference27th ACM SIGSPATIAL International Conference on Advances in Geographic Information Systems
Location350 W Mart Center Dr
CountryUnited States
CityChicago
Period05/11/201908/11/2019
Internet address

Keywords

  • Road Network
  • Machine Learning
  • Graph Representation Learning
  • Graph Convolutional Networks
  • Transportation

Fingerprint Dive into the research topics of 'Graph Convolutional Networks for Road Networks'. Together they form a unique fingerprint.

Cite this