The potential of explainable deep learning for public transport planning

Wenzhe Sun, Jan-Dirk Schmöcker, Youxi Lai and Koji Fukuda

Chapter 6 in Handbook on Artificial Intelligence and Transport, 2023, pp 155-175 from Edward Elgar Publishing

Abstract: Deep learning has been widely used in transportation studies, especially for prediction tasks. It has been proven powerful in predicting travel demand for a transportation node or a general spatial unit. Recently, graph convolutional networks (GCNs) have drawn much attention with their ability to model complex interactions between spatial units of an area by describing them with graphs. Multi-relational GCNs can be used for cases that consider multiple node features and relations between nodes. These have a variety of applications for public transport planning. To introduce GCNs and their variants to a wider range of audiences, this chapter aims to provide a general and easy-to-follow framework based on several examples. Furthermore, this chapter discusses the potential to explain such black boxes by using Shapley additive explanations (SHAP). SHAP may address the concern that arises among transport planners and operators regarding model interpretability.

Keywords: Economics and Finance; Environment; Geography; Innovations and Technology; Law - Academic; Politics and Public Policy Urban and Regional Studies (search for similar items in EconPapers)
Date: 2023
References: Add references at CitEc
Citations:

Downloads: (external link)
https://www.elgaronline.com/doi/10.4337/9781803929545.00013 (application/pdf)
Our link check indicates that this URL is bad, the error code is: 503 Service Temporarily Unavailable

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:elg:eechap:21868_6

Ordering information: This item can be ordered from
http://www.e-elgar.com

Access Statistics for this chapter

More chapters in Chapters from Edward Elgar Publishing
Bibliographic data for series maintained by Darrel McCalla ().