Advanced search
1 file | 1.43 MB Add to list

Predicting train occupancies based on query logs and external data sources

Gilles Vandewiele (UGent) , Joachim Van Herwegen (UGent) , Femke Ongenae (UGent) , Pieter Colpaert (UGent) , Ruben Verborgh (UGent) , Filip De Turck (UGent) , Olivier Janssens (UGent) and Erik Mannens (UGent)
Author
Organization
Abstract
On dense railway networks -such as in Belgium-train travelers are frequently confronted with overly occupied trains, especially during peak hours. Crowdedness on trains leads to a deterioration in the quality of service and has a negative impact on the well-being of the passenger. In order to stimulate travelers to consider less crowded trains, the iRail project wants to show an occupancy indicator in their route planning applications by the means of predictive modeling. As there is no official occupancy data available, training data is obtained by crowd-sourcing using the iRail web app(1) and the mobile Railer application for iPhone(2). Users can indicate their departure & arrival station, at what time they took a train and classify the occupancy of that train into the classes: low, medium or high. While preliminary results on a limited dataset conclude that the models do not yet perform sufficiently well, we are convinced that with further research and a larger amount of data, our predictive model will be able to achieve higher predictive performances. All datasets used in the current research are, for that purpose, made publicly available under an open license on the iRail website(3) and in the form of a Kaggle competition(4). Moreover, an infrastructure is set up that automatically processes new logs submitted by users in order for our model to continuously learn. Occupancy predictions for future trains are made available through an api(5).
Keywords
IBCN, predictive modeling, public transport, linked data

Downloads

  • 6849.pdf
    • full text (Published version)
    • |
    • open access
    • |
    • PDF
    • |
    • 1.43 MB

Citation

Please use this url to cite or link to this publication:

MLA
Vandewiele, Gilles, et al. “Predicting Train Occupancies Based on Query Logs and External Data Sources.” WWW’17 COMPANION : PROCEEDINGS OF THE 26TH INTERNATIONAL CONFERENCE ON WORLD WIDE WEB, 2017, pp. 1469–74.
APA
Vandewiele, G., Van Herwegen, J., Ongenae, F., Colpaert, P., Verborgh, R., De Turck, F., … Mannens, E. (2017). Predicting train occupancies based on query logs and external data sources. WWW’17 COMPANION : PROCEEDINGS OF THE 26TH INTERNATIONAL CONFERENCE ON WORLD WIDE WEB, 1469–1474.
Chicago author-date
Vandewiele, Gilles, Joachim Van Herwegen, Femke Ongenae, Pieter Colpaert, Ruben Verborgh, Filip De Turck, Olivier Janssens, and Erik Mannens. 2017. “Predicting Train Occupancies Based on Query Logs and External Data Sources.” In WWW’17 COMPANION : PROCEEDINGS OF THE 26TH INTERNATIONAL CONFERENCE ON WORLD WIDE WEB, 1469–74.
Chicago author-date (all authors)
Vandewiele, Gilles, Joachim Van Herwegen, Femke Ongenae, Pieter Colpaert, Ruben Verborgh, Filip De Turck, Olivier Janssens, and Erik Mannens. 2017. “Predicting Train Occupancies Based on Query Logs and External Data Sources.” In WWW’17 COMPANION : PROCEEDINGS OF THE 26TH INTERNATIONAL CONFERENCE ON WORLD WIDE WEB, 1469–1474.
Vancouver
1.
Vandewiele G, Van Herwegen J, Ongenae F, Colpaert P, Verborgh R, De Turck F, et al. Predicting train occupancies based on query logs and external data sources. In: WWW’17 COMPANION : PROCEEDINGS OF THE 26TH INTERNATIONAL CONFERENCE ON WORLD WIDE WEB. 2017. p. 1469–74.
IEEE
[1]
G. Vandewiele et al., “Predicting train occupancies based on query logs and external data sources,” in WWW’17 COMPANION : PROCEEDINGS OF THE 26TH INTERNATIONAL CONFERENCE ON WORLD WIDE WEB, Perth, Australia, 2017, pp. 1469–1474.
@inproceedings{8518282,
  abstract     = {{On dense railway networks -such as in Belgium-train travelers are frequently confronted with overly occupied trains, especially during peak hours. Crowdedness on trains leads to a deterioration in the quality of service and has a negative impact on the well-being of the passenger. In order to stimulate travelers to consider less crowded trains, the iRail project wants to show an occupancy indicator in their route planning applications by the means of predictive modeling. As there is no official occupancy data available, training data is obtained by crowd-sourcing using the iRail web app(1) and the mobile Railer application for iPhone(2). Users can indicate their departure & arrival station, at what time they took a train and classify the occupancy of that train into the classes: low, medium or high. While preliminary results on a limited dataset conclude that the models do not yet perform sufficiently well, we are convinced that with further research and a larger amount of data, our predictive model will be able to achieve higher predictive performances. All datasets used in the current research are, for that purpose, made publicly available under an open license on the iRail website(3) and in the form of a Kaggle competition(4). Moreover, an infrastructure is set up that automatically processes new logs submitted by users in order for our model to continuously learn. Occupancy predictions for future trains are made available through an api(5).}},
  author       = {{Vandewiele, Gilles and Van Herwegen, Joachim and Ongenae, Femke and Colpaert, Pieter and Verborgh, Ruben and De Turck, Filip and Janssens, Olivier and Mannens, Erik}},
  booktitle    = {{WWW'17 COMPANION : PROCEEDINGS OF THE 26TH INTERNATIONAL CONFERENCE ON WORLD WIDE WEB}},
  isbn         = {{9781450349147}},
  keywords     = {{IBCN,predictive modeling,public transport,linked data}},
  language     = {{eng}},
  location     = {{Perth, Australia}},
  pages        = {{1469--1474}},
  title        = {{Predicting train occupancies based on query logs and external data sources}},
  year         = {{2017}},
}

Web of Science
Times cited: