SH_Traffic

Citation Author(s):
Lingbai
Kong
Hanchen
Yang
Wengen
Li
Yichao
Zhang
Jihong
Guan
Shuigeng
Zhou
Submitted by:
Lingbai Kong
Last updated:
Fri, 01/03/2025 - 09:55
DOI:
10.21227/jrg0-vb50
Data Format:
*.csv
License:
Creative Commons Attribution
74 Views
Categories:
Machine Learning
Transportation
Keywords:
Shanghai, Spatio-Temporal Data Mining, Traffic flow forecasting, Graph convolution network, Deep Learning
0
0 ratings - Please login to submit your rating.
ACCESS DATASET CITE SHARE/EMBED

Abstract 

The SH_traffic dataset presents comprehensive traffic speed data for the Shanghai urban area, collected over a period from January 25, 2022, to February 9, 2022. The dataset encompasses a square area of 900 km² and consists of 4,500 graph nodes representing road intersections across the city. The dataset includes traffic speed information recorded at regular intervals, along with related data such as geographical entity attributes and relationships between entities within the road network.

Instructions: 

The SH_traffic dataset is organized into three CSV files and one JSON file following the standards of Atomic Files from LibCity: An Open Library for Urban Spatial-temporal Data Mining. The CSV files contain:

  • SH_traffic.rel: Relationship information between entities, such as road networks. Each entry includes a unique relationship ID, type, origin and destination IDs, road name, direction, and weight.

  • SH_traffic.dyna: Traffic condition information including a unique dynamic ID, type, timestamp, entity ID, and traffic speed data.

  • SH_traffic.geo: Geographic entity attribute information, including a unique geographic ID, type, coordinates, and various attribute categories such as Food, Shopping, Medical, and others.

The accompanying config.json file provides supplementary descriptions and configuration information for the dataset.

Dataset Files

LOGIN TO ACCESS DATASET FILES

 

QUESTIONS?