Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf
BRIGHT - Drift-Aware Demand Predictions for Taxi Networks
Show others and affiliations
2018 (English)In: IEEE Transactions on Knowledge and Data Engineering, ISSN 1041-4347, E-ISSN 1558-2191Article in journal (Refereed) Published
Abstract [en]

Massive data broadcast by GPS-equipped vehicles provide unprecedented opportunities. One of the main tasks in order to optimize our transportation networks is to build data-driven real-time decision support systems. However, the dynamic environments where the networks operate disallow the traditional assumptions required to put in practice many off-the-shelf supervised learning algorithms, such as finite training sets or stationary distributions. In this paper, we propose BRIGHT: a drift-aware supervised learning framework to predict demand quantities. BRIGHT aims to provide accurate predictions for short-term horizons through a creative ensemble of time series analysis methods that handles distinct types of concept drift. By selecting neighborhoods dynamically, BRIGHT reduces the likelihood of overfitting. By ensuring diversity among the base learners, BRIGHT ensures a high reduction of variance while keeping bias stable. Experiments were conducted using three large-scale heterogeneous real-world transportation networks in Porto (Portugal), Shanghai (China) and Stockholm (Sweden), as well as controlled experiments using synthetic data where multiple distinct drifts were artificially induced. The obtained results illustrate the advantages of BRIGHT in relation to state-of-the-art methods for this task. 

Place, publisher, year, edition, pages
IEEE, 2018.
Keywords [en]
Time-series forecasting; concept drift; ensemble learning; global positioning system (GPS) data; mobility intelligence; taxi passenger demand; machine learning
National Category
Transport Systems and Logistics
Identifiers
URN: urn:nbn:se:kth:diva-258021DOI: 10.1109/TKDE.2018.2883616ISI: 000507883700003Scopus ID: 2-s2.0-85058130950OAI: oai:DiVA.org:kth-258021DiVA, id: diva2:1349655
Funder
TrenOp, Transport Research Environment with Novel Perspectives
Note

QC 20191028

Available from: 2019-09-09 Created: 2019-09-09 Last updated: 2020-02-17Bibliographically approved

Open Access in DiVA

No full text in DiVA

Other links

Publisher's full textScopus

Authority records BETA

Jenelius, Erik

Search in DiVA

By author/editor
Jenelius, Erik
By organisation
Centre for Transport Studies, CTSTransport planning
In the same journal
IEEE Transactions on Knowledge and Data Engineering
Transport Systems and Logistics

Search outside of DiVA

GoogleGoogle Scholar

doi
urn-nbn

Altmetric score

doi
urn-nbn
Total: 86 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf