Spatio-Temporal Graph Convolutional Networks: A Deep Learning Framework for Traffic Forecasting

Bing Yu, Haoteng Yin, Zhanxing Zhu
9/14/2017
cs.LGstat.ML

Official code: VeritasYin/STGCN_IJCAI-18 (1,243 stars, BSD-2-Clause, last commit 2026-06-14).

Abstract

Timely accurate traffic forecast is crucial for urban traffic control and guidance. Due to the high nonlinearity and complexity of traffic flow, traditional methods cannot satisfy the requirements of mid-and-long term prediction tasks and often neglect spatial and temporal dependencies. In this paper, we propose a novel deep learning framework, Spatio-Temporal Graph Convolutional Networks (STGCN), to tackle the time series prediction problem in traffic domain. Instead of applying regular convolutional and recurrent units, we formulate the problem on graphs and build the model with complete convolutional structures, which enable much faster training speed with fewer parameters. Experiments show that our model STGCN effectively captures comprehensive spatio-temporal correlations through modeling multi-scale traffic networks and consistently outperforms state-of-the-art baselines on various real-world traffic datasets.

View on arXivView PDF

Code Implementations(1)

VeritasYin/STGCN_IJCAI-18Official100%
BSD-2-Clause

[IJCAI'18] Spatio-Temporal Graph Convolutional Networks

1,243326PythonApr 15, 20184 months agoBSD-2-Clause

Similar Papers

T-GCN: A Temporal Graph ConvolutionalNetwork for Traffic Prediction

91% match1 repo1.8k2018

Traffic Graph Convolutional Recurrent Neural Network: A Deep Learning Framework for Network-Scale Traffic Learning and Forecasting

91% match1 repo982018

Enhancing Spatiotemporal Traffic Prediction through Urban Human Activity Analysis

90% match1 repo82023

Causally-Aware Spatio-Temporal Multi-Graph Convolution Network for Accurate and Reliable Traffic Prediction

90% match1 repo22024

DL-Traff: Survey and Benchmark of Deep Learning Models for Urban Traffic Prediction

90% match2 repos1832021

DKGCM: A Spatio-Temporal Prediction Model for Traffic Flow by Fusing Spatial Node Clustering Method and Fourier Bidirectional Mamba Mechanism

90% match2025

Cite this paper

@article{yu2017spatiotemporal,
  title  = {Spatio-Temporal Graph Convolutional Networks: A Deep Learning Framework for Traffic Forecasting},
  author = {Bing Yu and Haoteng Yin and Zhanxing Zhu},
  year   = {2017},
  eprint = {1709.04875},
  archivePrefix = {arXiv},
  url    = {https://arxiv.org/abs/1709.04875}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/1709.04875.svg)](https://codeofpaper.com/paper/1709.04875)

Discussion