EUVIMEDEUVIMED entdecken
Crossref

Cross-City Pretraining Transfer Learning Model for Traffic Flow Prediction

Zhizhe Lin, Zequan Li, Chaozhi Yu, Chunjie Cao, Teng Zhou, Guangyin Jin

IEEE Internet of Things Journal · 01.07.2026 · 13 (13) · 28707-28717

DOI: 10.1109/jiot.2026.3685259 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Internet of Things Journal, erschienen am 01.07.2026. Verfasst von Zhizhe Lin und 5 weiteren Autor:innen. Die Arbeit wurde bisher 7-mal zitiert (Crossref) und nennt 62 Literaturangaben. Gefördert von: National Natural Science Foundation of China, Guangxi Natural Science Foundation, Philosophy and Social Sciences Planning Project of Zhejiang Province.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Accurate traffic flow prediction plays a pivotal role in intelligent transportation systems (ITSs). While deep learning-based approaches have demonstrated remarkable success in this domain, their performance heavily depends on the availability of large-scale training data. However, many cities face challenges in collecting sufficient traffic flow data due to privacy concerns and substantial storage requirements. Consequently, conventional traffic flow prediction models often suffer from performance degradation when applied to cities with limited data availability, primarily due to spatially unbalanced data distributions. To overcome this limitation, we propose a novel pretrained framework for cross-city traffic flow prediction, termed pretraining cross-city (PTCC). Different from existing methods that focus solely on optimizing performance for data-rich cities, our framework innovatively transfers spatiotemporal knowledge from data-abundant cities to enhance prediction accuracy in data-scarce scenarios. The proposed PTCC framework comprises three key components: 1) a pretrained module that learns long-term temporal patterns from traffic flow data in source cities and generates comprehensive segment-level representations; 2) a discrete graph learning structure that captures node dependencies from contextual segment-level representations; and 3) a spatiotemporal prediction module that effectively transfers the acquired knowledge to facilitate accurate traffic flow forecasting in target cities. We conduct extensive experiments to validate the framework’s effectiveness, training the model on METR-LA and PEMS-BAY datasets, and evaluating its performance on the PEMS04 and PEMS08 datasets. The experimental results demonstrate that our pretrained framework significantly outperforms existing methods, establishing its superiority for traffic flow prediction in cities with limited data availability.

Förderung

  • National Natural Science Foundation of China
  • Guangxi Natural Science Foundation
  • Philosophy and Social Sciences Planning Project of Zhejiang Province
  • Hainan Provincial Natural Science Foundation of China
  • Basic and Applied Basic Research Foundation of Guangdong Province

Zitieren

APA

Lin, Z., Li, Z., Yu, C., Cao, C., Zhou, T., & Jin, G. (2026). Cross-City Pretraining Transfer Learning Model for Traffic Flow Prediction. IEEE Internet of Things Journal, 13(13), 28707-28717. https://doi.org/10.1109/jiot.2026.3685259

Vancouver

Lin Z, Li Z, Yu C, Cao C, Zhou T, Jin G. Cross-City Pretraining Transfer Learning Model for Traffic Flow Prediction. IEEE Internet of Things Journal. 2026;13(13):28707-28717. doi:10.1109/jiot.2026.3685259

BibTeX
@article{Lin2026,
  author = {Lin, Zhizhe and Li, Zequan and Yu, Chaozhi and Cao, Chunjie and Zhou, Teng and Jin, Guangyin},
  title = {Cross-City Pretraining Transfer Learning Model for Traffic Flow Prediction},
  journal = {IEEE Internet of Things Journal},
  year = {2026},
  volume = {13},
  number = {13},
  pages = {28707--28717},
  doi = {10.1109/jiot.2026.3685259}
}
Kostenlos starten →