Dynamic Graph Convolution Network with Multi-head Attention for Traffic Flow Prediction
Purposes Traffic flow prediction is crucial for the effective management and operation of urban transportation systems. The flows of different road sections or intersections in a traffic network change dynamically with time, meanwhile the flows of spatially neighboring road sections or intersections...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Editorial Office of Journal of Taiyuan University of Technology
2024-01-01
|
| Series: | Taiyuan Ligong Daxue xuebao |
| Subjects: | |
| Online Access: | https://tyutjournal.tyut.edu.cn/englishpaper/show-2257.html |
| _version_ | 1797208289853833216 |
|---|---|
| author | Hanyou DENG Hongmei CHEN Qing XIAO Yuan FANG |
| author_facet | Hanyou DENG Hongmei CHEN Qing XIAO Yuan FANG |
| author_sort | Hanyou DENG |
| collection | DOAJ |
| description | Purposes Traffic flow prediction is crucial for the effective management and operation of urban transportation systems. The flows of different road sections or intersections in a traffic network change dynamically with time, meanwhile the flows of spatially neighboring road sections or intersections affect each other. In order to better learn the spatial and temporal correlation of the traffic flow of different road sections or intersections from the traffic flow sequences, and to improve the performance of short-term prediction of traffic flow, in this paper we propose a traffic flow prediction method based on Dynamic Graph Convolution Network with Multi-head Attention (DGCNMA). Methods The DGCNMA model first introduces graph convolution networks into the Transformer framework to learn the spatial embedding of traffic flow sequences and incorporate them into the traffic flow sequences, and then adopts the mechanism of Multi-head Attention to capture the temporal and spatial correlation of the traffic flow sequences from multiple perspectives at the same time; second, the Interactive Dynamic Graph Convolution Network is introduced to simultaneously learn the local and global spatial-temporal correlations of traffic flow sequences through the interactive learning of convolutional network and dynamic graph convolutional network, and the interactive fusion of parity subsequence features. Findings Experiments on highway traffic flow datasets (PEMS03, PEMS04, PEMS08) and subway crowd flow datasets (HZME inflow and HZME outflow) show that the proposed DGCNMA model has better traffic flow prediction performance than the baseline models. |
| first_indexed | 2024-04-24T09:36:27Z |
| format | Article |
| id | doaj.art-6cf03a7633d24ab6b1d7a9307c447a8e |
| institution | Directory Open Access Journal |
| issn | 1007-9432 |
| language | English |
| last_indexed | 2024-04-24T09:36:27Z |
| publishDate | 2024-01-01 |
| publisher | Editorial Office of Journal of Taiyuan University of Technology |
| record_format | Article |
| series | Taiyuan Ligong Daxue xuebao |
| spelling | doaj.art-6cf03a7633d24ab6b1d7a9307c447a8e2024-04-15T09:17:22ZengEditorial Office of Journal of Taiyuan University of TechnologyTaiyuan Ligong Daxue xuebao1007-94322024-01-0155117218310.16355/j.tyut.1007-9432.2023BD0041007-9432(2024)01-0172-12Dynamic Graph Convolution Network with Multi-head Attention for Traffic Flow PredictionHanyou DENG0Hongmei CHEN1Qing XIAO2Yuan FANG3School of Information Science and Engineering, Yunnan University, Kunming 650500, ChinaSchool of Information Science and Engineering, Yunnan University, Kunming 650500, ChinaSchool of Information Science and Engineering, Yunnan University, Kunming 650500, ChinaSouth-Western Institute for Astronomy Research, Yunnan University, Kunming 650500, ChinaPurposes Traffic flow prediction is crucial for the effective management and operation of urban transportation systems. The flows of different road sections or intersections in a traffic network change dynamically with time, meanwhile the flows of spatially neighboring road sections or intersections affect each other. In order to better learn the spatial and temporal correlation of the traffic flow of different road sections or intersections from the traffic flow sequences, and to improve the performance of short-term prediction of traffic flow, in this paper we propose a traffic flow prediction method based on Dynamic Graph Convolution Network with Multi-head Attention (DGCNMA). Methods The DGCNMA model first introduces graph convolution networks into the Transformer framework to learn the spatial embedding of traffic flow sequences and incorporate them into the traffic flow sequences, and then adopts the mechanism of Multi-head Attention to capture the temporal and spatial correlation of the traffic flow sequences from multiple perspectives at the same time; second, the Interactive Dynamic Graph Convolution Network is introduced to simultaneously learn the local and global spatial-temporal correlations of traffic flow sequences through the interactive learning of convolutional network and dynamic graph convolutional network, and the interactive fusion of parity subsequence features. Findings Experiments on highway traffic flow datasets (PEMS03, PEMS04, PEMS08) and subway crowd flow datasets (HZME inflow and HZME outflow) show that the proposed DGCNMA model has better traffic flow prediction performance than the baseline models.https://tyutjournal.tyut.edu.cn/englishpaper/show-2257.htmltraffic flow predictionmulti-head attentioninteractive dynamic graph convolution network |
| spellingShingle | Hanyou DENG Hongmei CHEN Qing XIAO Yuan FANG Dynamic Graph Convolution Network with Multi-head Attention for Traffic Flow Prediction Taiyuan Ligong Daxue xuebao traffic flow prediction multi-head attention interactive dynamic graph convolution network |
| title | Dynamic Graph Convolution Network with Multi-head Attention for Traffic Flow Prediction |
| title_full | Dynamic Graph Convolution Network with Multi-head Attention for Traffic Flow Prediction |
| title_fullStr | Dynamic Graph Convolution Network with Multi-head Attention for Traffic Flow Prediction |
| title_full_unstemmed | Dynamic Graph Convolution Network with Multi-head Attention for Traffic Flow Prediction |
| title_short | Dynamic Graph Convolution Network with Multi-head Attention for Traffic Flow Prediction |
| title_sort | dynamic graph convolution network with multi head attention for traffic flow prediction |
| topic | traffic flow prediction multi-head attention interactive dynamic graph convolution network |
| url | https://tyutjournal.tyut.edu.cn/englishpaper/show-2257.html |
| work_keys_str_mv | AT hanyoudeng dynamicgraphconvolutionnetworkwithmultiheadattentionfortrafficflowprediction AT hongmeichen dynamicgraphconvolutionnetworkwithmultiheadattentionfortrafficflowprediction AT qingxiao dynamicgraphconvolutionnetworkwithmultiheadattentionfortrafficflowprediction AT yuanfang dynamicgraphconvolutionnetworkwithmultiheadattentionfortrafficflowprediction |