Multi-featured spatial-temporal and dynamic multi-graph convolutional network for metro passenger flow prediction

Metro passenger flow prediction is an essential part of crowd flow forecasting and intelligent transportation management systems. However, two challenges still need to be addressed to achieve a more accurate prediction: (1) accounting for featural dependence instead of considering only the temporal...

Full description

Saved in:
Bibliographic Details
Main Authors: Chuan Zhao, Xin Li, Zezhi Shao, HongJi Yang, Fei Wang
Format: Article
Language:English
Published: Taylor & Francis Group 2022-12-01
Series:Connection Science
Subjects:
Online Access:http://dx.doi.org/10.1080/09540091.2022.2061915
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849684217673285632
author Chuan Zhao
Xin Li
Zezhi Shao
HongJi Yang
Fei Wang
author_facet Chuan Zhao
Xin Li
Zezhi Shao
HongJi Yang
Fei Wang
author_sort Chuan Zhao
collection DOAJ
description Metro passenger flow prediction is an essential part of crowd flow forecasting and intelligent transportation management systems. However, two challenges still need to be addressed to achieve a more accurate prediction: (1) accounting for featural dependence instead of considering only the temporal connection and spatial relations; (2) utilising graph structures to address non-European relationships of spatial and featural dependence. To address these challenges, we developed a novel model called the multi-featured spatial-temporal (MFST) and dynamic multi-graph convolutional network (DMGCN) model. Temporal connections are learned from both the local and global information in a time-series sequence using the combination of a time-trend feature mapping block and a gated recurrent unit block. Spatial relation and featural dependence are separately captured by two DMGCN blocks. Each DMGCN block encodes various relationships by constructing multiple graphs consisting of predefined and non-defined topologies. The results of evaluations conducted of the MFST tensor and the DMGCN on the real-world Beijing subway dataset indicate that the prediction performance of the proposed model is superior to that of the existing baselines. The proposed model thus contributes significantly to the improvement of public safety by providing early warnings of large passenger flow and enabling the smart scheduling of resources.
format Article
id doaj-art-bd425c77c9e64a79b139f3da9a2233cb
institution DOAJ
issn 0954-0091
1360-0494
language English
publishDate 2022-12-01
publisher Taylor & Francis Group
record_format Article
series Connection Science
spelling doaj-art-bd425c77c9e64a79b139f3da9a2233cb2025-08-20T03:23:31ZengTaylor & Francis GroupConnection Science0954-00911360-04942022-12-013411252127210.1080/09540091.2022.20619152061915Multi-featured spatial-temporal and dynamic multi-graph convolutional network for metro passenger flow predictionChuan Zhao0Xin Li1Zezhi Shao2HongJi Yang3Fei Wang4Beijing Technology and Business UniversityBeijing Technology and Business UniversityChinese Academy of SciencesUniversity of LeicesterChinese Academy of SciencesMetro passenger flow prediction is an essential part of crowd flow forecasting and intelligent transportation management systems. However, two challenges still need to be addressed to achieve a more accurate prediction: (1) accounting for featural dependence instead of considering only the temporal connection and spatial relations; (2) utilising graph structures to address non-European relationships of spatial and featural dependence. To address these challenges, we developed a novel model called the multi-featured spatial-temporal (MFST) and dynamic multi-graph convolutional network (DMGCN) model. Temporal connections are learned from both the local and global information in a time-series sequence using the combination of a time-trend feature mapping block and a gated recurrent unit block. Spatial relation and featural dependence are separately captured by two DMGCN blocks. Each DMGCN block encodes various relationships by constructing multiple graphs consisting of predefined and non-defined topologies. The results of evaluations conducted of the MFST tensor and the DMGCN on the real-world Beijing subway dataset indicate that the prediction performance of the proposed model is superior to that of the existing baselines. The proposed model thus contributes significantly to the improvement of public safety by providing early warnings of large passenger flow and enabling the smart scheduling of resources.http://dx.doi.org/10.1080/09540091.2022.2061915metro passenger flow predictiondeep learningmulti-featured spatial-temporal tensordynamic multi-graph neural network
spellingShingle Chuan Zhao
Xin Li
Zezhi Shao
HongJi Yang
Fei Wang
Multi-featured spatial-temporal and dynamic multi-graph convolutional network for metro passenger flow prediction
Connection Science
metro passenger flow prediction
deep learning
multi-featured spatial-temporal tensor
dynamic multi-graph neural network
title Multi-featured spatial-temporal and dynamic multi-graph convolutional network for metro passenger flow prediction
title_full Multi-featured spatial-temporal and dynamic multi-graph convolutional network for metro passenger flow prediction
title_fullStr Multi-featured spatial-temporal and dynamic multi-graph convolutional network for metro passenger flow prediction
title_full_unstemmed Multi-featured spatial-temporal and dynamic multi-graph convolutional network for metro passenger flow prediction
title_short Multi-featured spatial-temporal and dynamic multi-graph convolutional network for metro passenger flow prediction
title_sort multi featured spatial temporal and dynamic multi graph convolutional network for metro passenger flow prediction
topic metro passenger flow prediction
deep learning
multi-featured spatial-temporal tensor
dynamic multi-graph neural network
url http://dx.doi.org/10.1080/09540091.2022.2061915
work_keys_str_mv AT chuanzhao multifeaturedspatialtemporalanddynamicmultigraphconvolutionalnetworkformetropassengerflowprediction
AT xinli multifeaturedspatialtemporalanddynamicmultigraphconvolutionalnetworkformetropassengerflowprediction
AT zezhishao multifeaturedspatialtemporalanddynamicmultigraphconvolutionalnetworkformetropassengerflowprediction
AT hongjiyang multifeaturedspatialtemporalanddynamicmultigraphconvolutionalnetworkformetropassengerflowprediction
AT feiwang multifeaturedspatialtemporalanddynamicmultigraphconvolutionalnetworkformetropassengerflowprediction