2020
DOI: 10.1109/access.2020.3018452
|View full text |Cite
|
Sign up to set email alerts
|

A General Traffic Flow Prediction Approach Based on Spatial-Temporal Graph Attention

Abstract: Accurate and reliable traffic flow prediction is critical to the safe and stable deployment of intelligent transportation systems. However, it is very challenging since the complex spatial and temporal dependence of traffic flows. Most existing works require the information of the traffic network structure and human intervention to model the spatial-temporal association of traffic data, resulting in low generality of the model and unsatisfactory prediction performance. In this paper, we propose a general spati… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1
1

Citation Types

0
10
0

Year Published

2021
2021
2023
2023

Publication Types

Select...
8
2

Relationship

0
10

Authors

Journals

citations
Cited by 29 publications
(10 citation statements)
references
References 32 publications
0
10
0
Order By: Relevance
“…The attention-based learning methods use the attention mechanism to learn the pairwise inter-variable dependencies. For traffic flow forecasting, Tang et al [31] used a graph attention module to learn graph structure. Zheng et al [33] used spatial attention mechanism to learn the correlation of traffic flow at different nodes.…”
Section: B Graph Learning For Mtsmentioning
confidence: 99%
“…The attention-based learning methods use the attention mechanism to learn the pairwise inter-variable dependencies. For traffic flow forecasting, Tang et al [31] used a graph attention module to learn graph structure. Zheng et al [33] used spatial attention mechanism to learn the correlation of traffic flow at different nodes.…”
Section: B Graph Learning For Mtsmentioning
confidence: 99%
“…As is shown by ( Figure 1 ), the sensors deployed on the traffic network which collected traffic data at fixed time intervals form a non-Euclidean topological graph naturally ( 34 ). We use nodes to represent the locations of traffic sensors, and the road segments connecting traffic sensors are treated as edges in graph.…”
Section: Problem Descriptionmentioning
confidence: 99%
“…Flows, Crowds and Density. Some existing works study the estimation of flows [33,34], crowds [6,37], or dense regions [15,16] outdoors. However, they all fall short in indoor spaces mainly for two reasons.…”
Section: Related Workmentioning
confidence: 99%