Traffic flow prediction based on a multi-task pivotal attention network
PeerJ Computer Science · 30.06.2026 · 12 · e3924
DOI: 10.7717/peerj-cs.3924 ↗Kurz gefasst
Fachartikel in der Zeitschrift PeerJ Computer Science, online erschienen am 30.06.2026. Verfasst von Qiaoli Liu und 1 weiteren Autor:innen. Die Arbeit nennt 43 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: “Sanmenxia 2024 Science and Technology Development Program Project.
Abstract (Originalsprache)
Short-term traffic flow prediction remains challenging because traffic dynamics are nonlinear, non-stationary, and governed by both local topology and long-range directional dependencies. Existing approaches often rely on static or uniformly learned graphs, do not explicitly separate slow trends from rapid fluctuations, and insufficiently exploit correlated modalities such as flow, speed, and traffic type. To address these issues, we propose the Pivotal Multi-Modal Spatio-Temporal Network (PMST-Net), which combines a Pivotal Node Identification Module (PNIM), a Temporal Feature Disentanglement (TFD) layer, a Pivotal-Informed Spatial Attention (PISA) encoder, and a Cross-Modal Multi-Task Learning (CM-MTL) prediction head. Experiments on four public PeMS benchmarks demonstrate consistent improvements over strong baselines. On PEMS08, PMST-Net reduces mean absolute error (MAE) and root mean squared error (RMSE) by 42.3% and 37.2%, respectively, relative to the strongest baseline. Ablation, robustness, and efficiency analyses further confirm the contribution of each module and the practical value of the overall design.
Förderung
- “Sanmenxia 2024 Science and Technology Development Program Project
Zitieren
APA
Liu, Q., & Zhang, Y. (2026). Traffic flow prediction based on a multi-task pivotal attention network. PeerJ Computer Science, 12, e3924. https://doi.org/10.7717/peerj-cs.3924
Vancouver
Liu Q, Zhang Y. Traffic flow prediction based on a multi-task pivotal attention network. PeerJ Computer Science. 2026;12:e3924. doi:10.7717/peerj-cs.3924
BibTeX
@article{Liu2026,
author = {Liu, Qiaoli and Zhang, Yali},
title = {Traffic flow prediction based on a multi-task pivotal attention network},
journal = {PeerJ Computer Science},
year = {2026},
volume = {12},
pages = {e3924},
doi = {10.7717/peerj-cs.3924}
}
EUVIMED