Graph convolutional network approach applied to predict hourly bike-sharing demands considering spatial, temporal, and global effects

Tae San Kim, Won Kyung Lee, So Young Sohn

Research output: Contribution to journalArticle

Abstract

Solving the supply–demand imbalance is the most crucial issue for stable implementation of a public bike-sharing system. This gap can be reduced by increasing the accuracy of demand prediction by considering spatial and temporal properties of bike demand. However, only a few attempts have been made to account for both features simultaneously. Therefore, we propose a prediction framework based on graph convolutional networks. Our framework reflects not only spatial dependencies among stations, but also various temporal patterns over different periods. Additionally, we consider the influence of global variables, such as weather and weekday/weekend to reflect non-station-level changes. We compare our framework to other baseline models using the data from Seoul’s bike-sharing system. Results show that our approach has better performance than existing prediction models.

Original languageEnglish
Article numbere0220782
JournalPloS one
Volume14
Issue number9
DOIs
Publication statusPublished - 2019 Jan 1

Fingerprint

Spatial Analysis
prediction
Weather
weather

All Science Journal Classification (ASJC) codes

  • Biochemistry, Genetics and Molecular Biology(all)
  • Agricultural and Biological Sciences(all)
  • General

Cite this

@article{09a30cfd5cfe48a49c45c325eec024bc,
title = "Graph convolutional network approach applied to predict hourly bike-sharing demands considering spatial, temporal, and global effects",
abstract = "Solving the supply–demand imbalance is the most crucial issue for stable implementation of a public bike-sharing system. This gap can be reduced by increasing the accuracy of demand prediction by considering spatial and temporal properties of bike demand. However, only a few attempts have been made to account for both features simultaneously. Therefore, we propose a prediction framework based on graph convolutional networks. Our framework reflects not only spatial dependencies among stations, but also various temporal patterns over different periods. Additionally, we consider the influence of global variables, such as weather and weekday/weekend to reflect non-station-level changes. We compare our framework to other baseline models using the data from Seoul’s bike-sharing system. Results show that our approach has better performance than existing prediction models.",
author = "Kim, {Tae San} and Lee, {Won Kyung} and Sohn, {So Young}",
year = "2019",
month = "1",
day = "1",
doi = "10.1371/journal.pone.0220782",
language = "English",
volume = "14",
journal = "PLoS One",
issn = "1932-6203",
publisher = "Public Library of Science",
number = "9",

}

Graph convolutional network approach applied to predict hourly bike-sharing demands considering spatial, temporal, and global effects. / Kim, Tae San; Lee, Won Kyung; Sohn, So Young.

In: PloS one, Vol. 14, No. 9, e0220782, 01.01.2019.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Graph convolutional network approach applied to predict hourly bike-sharing demands considering spatial, temporal, and global effects

AU - Kim, Tae San

AU - Lee, Won Kyung

AU - Sohn, So Young

PY - 2019/1/1

Y1 - 2019/1/1

N2 - Solving the supply–demand imbalance is the most crucial issue for stable implementation of a public bike-sharing system. This gap can be reduced by increasing the accuracy of demand prediction by considering spatial and temporal properties of bike demand. However, only a few attempts have been made to account for both features simultaneously. Therefore, we propose a prediction framework based on graph convolutional networks. Our framework reflects not only spatial dependencies among stations, but also various temporal patterns over different periods. Additionally, we consider the influence of global variables, such as weather and weekday/weekend to reflect non-station-level changes. We compare our framework to other baseline models using the data from Seoul’s bike-sharing system. Results show that our approach has better performance than existing prediction models.

AB - Solving the supply–demand imbalance is the most crucial issue for stable implementation of a public bike-sharing system. This gap can be reduced by increasing the accuracy of demand prediction by considering spatial and temporal properties of bike demand. However, only a few attempts have been made to account for both features simultaneously. Therefore, we propose a prediction framework based on graph convolutional networks. Our framework reflects not only spatial dependencies among stations, but also various temporal patterns over different periods. Additionally, we consider the influence of global variables, such as weather and weekday/weekend to reflect non-station-level changes. We compare our framework to other baseline models using the data from Seoul’s bike-sharing system. Results show that our approach has better performance than existing prediction models.

UR - http://www.scopus.com/inward/record.url?scp=85072191368&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85072191368&partnerID=8YFLogxK

U2 - 10.1371/journal.pone.0220782

DO - 10.1371/journal.pone.0220782

M3 - Article

C2 - 31525227

AN - SCOPUS:85072191368

VL - 14

JO - PLoS One

JF - PLoS One

SN - 1932-6203

IS - 9

M1 - e0220782

ER -