Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

AMGST: Adaptive multi-graph convolution and spatiotemporal attention network for traffic forecasting.

Pei Shi1,2, Qixiang Lu3, Jiahui Chen1

  • 1School of IoT Engineering, Wuxi University, Wuxi, China.

Plos One
|June 4, 2026
PubMed
Summary

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Performance evaluation of equations of state for typical cryogens two-phase simulation by pseudopotential lattice Boltzmann method.

Physical review. E·2026
Same author

Lipopeptides and antibiotics from a marine <i>Bacillus pumilus</i> mediate a potential "catch and kill" effect on pathogenetic <i>Vibrio parahaemolyticus</i>.

mSystems·2025
Same author

Neuroprotective activity of thiolutin in epileptic mice: the inhibition of NLRP3 inflammasome.

Human cell·2025
Same author

An engineered antibody-lectin conjugate targeting the HIV glycan shield protects humanized mice against HIV challenge.

Molecular therapy : the journal of the American Society of Gene Therapy·2025
Same author

Lipoxin A4 modulates the PKA/CREB and NF-κB signaling pathway to mitigate chondrocyte catabolism and apoptosis in temporomandibular joint osteoarthritis.

Experimental cell research·2024
Same author

Diaphragmatic dysfunction is associated with postoperative pulmonary complications in the aged patients underwent radical resection of esophageal cancer: a prospective observational study.

Journal of thoracic disease·2024
Same journal

Invaders taking over-Mollusc faunal change in volcanic barrier lakes of the Albertine Rift biodiversity hotspot.

PloS one·2026
Same journal

AI-driven molecular diversification and ligand-based optimization of macitentan derivatives targeting VEGFR1 and endothelin signaling pathways.

PloS one·2026
Same journal

Performance patterns and records in the world aquatics masters championships: Where do the most frequently represented nations among the top-ten masters swimmers come from?

PloS one·2026
Same journal

Modeling diurnal Temperature-Rainfall relationships under multicollinearity using PLS-SEM: A case study of Ghana.

PloS one·2026
Same journal

Organizational culture, social capital, and emergency capacity in primary healthcare institutions: A cross-sectional structural equation modeling study comparing ordinary and older communities.

PloS one·2026
Same journal

Impact of kidney function on the metabolome in the general population.

PloS one·2026
See all related articles
This summary is machine-generated.

This study introduces AMGST, a novel network for traffic forecasting. It improves accuracy by better capturing spatial and temporal traffic patterns using adaptive multi-graph convolution and spatiotemporal attention.

Area of Science:

  • Artificial Intelligence
  • Computer Science
  • Transportation Engineering

Background:

  • Graph convolution shows promise for traffic prediction but struggles with global spatial correlations and long-term temporal dependencies.
  • The quality of graph structures significantly impacts the extraction of spatiotemporal traffic patterns.
  • Existing methods face challenges in fully capturing complex spatiotemporal dynamics in traffic data.

Purpose of the Study:

  • To propose AMGST, an Adaptive Multi-Graph Convolution and Spatiotemporal Multi-Head Self-Attention Network for enhanced traffic forecasting.
  • To address limitations in capturing global spatial correlations and long-term temporal dependencies in traffic prediction.
  • To improve the accuracy and robustness of traffic forecasting models.

Main Methods:

Related Experiment Videos

  • Developed an Adaptive Spatiotemporal Embedding (ASTE) generator for dynamic representations.
  • Implemented a multi-graph diffusion convolution using mutual information and adaptive matrices for fine-grained spatial feature extraction.
  • Integrated global spatial attention and temporal attention modules to capture dynamic spatial correlations and nonlinear temporal dependencies.

Main Results:

  • AMGST consistently outperformed baseline models across four public traffic datasets.
  • The model demonstrated effectiveness in predicting both traffic speed and flow.
  • Experimental results validate the proposed network's ability to capture complex spatiotemporal traffic dynamics.

Conclusions:

  • AMGST significantly enhances traffic forecasting accuracy by effectively modeling spatiotemporal dependencies.
  • The adaptive multi-graph convolution and spatiotemporal attention mechanisms are key to the model's superior performance.
  • The proposed approach offers a robust solution for intelligent traffic management systems.