Fast Spatiotemporal Learning Framework for Traffic Flow Forecasting

Guo, C; Chen, CH; Hwang, FJ; Chang, CC; Chang, CC

Fast Spatiotemporal Learning Framework for Traffic Flow Forecasting

Guo, C Chen, CH Hwang, FJ Chang, CC Chang, CC

Permalink

Publisher:: IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
Publication Type:: Journal Article
Citation:: IEEE Transactions on Intelligent Transportation Systems, 2023, 24, (8), pp. 8606-8616
Issue Date:: 2023

Closed Access

	Filename	Description	Size
	Fast_Spatiotemporal_Learning_Framework_for_Traffic_Flow_Forecasting.pdf	Published version	2.19 MB	Adobe PDF	View/Open

Copyright Clearance Process

Recently Added
In Progress
Closed Access

This item is closed access and not available.

Full metadata record

Field	Value	Language
dc.contributor.author	Guo, C
dc.contributor.author	Chen, CH
dc.contributor.author	Hwang, FJ
dc.contributor.author	Chang, CC
dc.contributor.author	Chang, CC
dc.date.accessioned	2024-03-19T01:56:02Z
dc.date.available	2024-03-19T01:56:02Z
dc.date.issued	2023
dc.identifier.citation	IEEE Transactions on Intelligent Transportation Systems, 2023, 24, (8), pp. 8606-8616
dc.identifier.issn	1524-9050
dc.identifier.issn	1558-0016
dc.identifier.uri	http://hdl.handle.net/10453/176894
dc.description.abstract	The graph convolution network (GCN), whose flexible convolution kernels perfectly adapt to the complex topology of the road network, has gradually dominated the spatiotemporal dependency learning of traffic flow data. Defining and learning the spatiotemporal characteristics and relationships of the traffic network efficiently and accurately, which are the important prerequisites for the success of the GCN, have become one of the most burning research problems in the field of intelligent transportation systems. This paper proposes a fast spatiotemporal learning (FSTL) framework containing the fast spatiotemporal GCN module, which reduces the computational complexity of the spatiotemporal GCN from <inline-formula> <tex-math notation="LaTeX">${{\cal O}(k^2)}$</tex-math> </inline-formula> to <inline-formula> <tex-math notation="LaTeX">${{\cal O}(k)}$</tex-math> </inline-formula>, where <inline-formula> <tex-math notation="LaTeX">$k$</tex-math> </inline-formula> is the number of time steps of data learned in each GCN operation. To mine globally and fast the correlations of road node pairs, a correlation analysis based on the normal distribution with the complexity of <inline-formula> <tex-math notation="LaTeX">${{\cal O}(N)}$</tex-math> </inline-formula>, where <inline-formula> <tex-math notation="LaTeX">$N$</tex-math> </inline-formula> is the number of nodes in the traffic network, is proposed to construct the global correlation matrix. Besides, the multi-scale temporal learning is integrated into the FSTL to overcome the receptive field constraints of the spatiotemporal GCN. The experimental results on four real-world datasets demonstrate that the FSTL achieves 48.88% and 5.26% reductions in the training time and mean absolute error, respectively, compared with the state-of-the-art model.
dc.language	English
dc.publisher	IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
dc.relation.ispartof	IEEE Transactions on Intelligent Transportation Systems
dc.relation.isbasedon	10.1109/TITS.2022.3224039
dc.rights	info:eu-repo/semantics/closedAccess
dc.subject	0801 Artificial Intelligence and Image Processing, 0905 Civil Engineering, 1507 Transportation and Freight Services
dc.subject.classification	Logistics & Transportation
dc.subject.classification	3509 Transportation, logistics and supply chains
dc.subject.classification	4602 Artificial intelligence
dc.subject.classification	4603 Computer vision and multimedia computation
dc.title	Fast Spatiotemporal Learning Framework for Traffic Flow Forecasting
dc.type	Journal Article
utslib.citation.volume	24
utslib.for	0801 Artificial Intelligence and Image Processing
utslib.for	0905 Civil Engineering
utslib.for	1507 Transportation and Freight Services
pubs.organisational-group	University of Technology Sydney
pubs.organisational-group	University of Technology Sydney/Faculty of Science
pubs.organisational-group	University of Technology Sydney/Faculty of Science/School of Mathematical and Physical Sciences
utslib.copyright.status	closed_access	*
pubs.consider-herdc	false
dc.date.updated	2024-03-19T01:56:01Z
pubs.issue	8
pubs.publication-status	Published
pubs.volume	24
utslib.citation.issue	8

Abstract:

The graph convolution network (GCN), whose flexible convolution kernels perfectly adapt to the complex topology of the road network, has gradually dominated the spatiotemporal dependency learning of traffic flow data. Defining and learning the spatiotemporal characteristics and relationships of the traffic network efficiently and accurately, which are the important prerequisites for the success of the GCN, have become one of the most burning research problems in the field of intelligent transportation systems. This paper proposes a fast spatiotemporal learning (FSTL) framework containing the fast spatiotemporal GCN module, which reduces the computational complexity of the spatiotemporal GCN from

${{\cal O}(k^2)}$

to

${{\cal O}(k)}$

, where

$k$

is the number of time steps of data learned in each GCN operation. To mine globally and fast the correlations of road node pairs, a correlation analysis based on the normal distribution with the complexity of

${{\cal O}(N)}$

, where

$N$

is the number of nodes in the traffic network, is proposed to construct the global correlation matrix. Besides, the multi-scale temporal learning is integrated into the FSTL to overcome the receptive field constraints of the spatiotemporal GCN. The experimental results on four real-world datasets demonstrate that the FSTL achieves 48.88% and 5.26% reductions in the training time and mean absolute error, respectively, compared with the state-of-the-art model.

Please use this identifier to cite or link to this item:

http://hdl.handle.net/10453/176894