A3T-GCN: Attention Temporal Graph Convolutional Network for Traffic Forecasting
Accurate real-time traffic forecasting is a core technological problem against the implementation of the intelligent transportation system. However, it remains challenging considering the complex spatial and temporal dependencies among traffic flows. In the spatial dimension, due to the connectivity...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2021-07-01
|
| Series: | ISPRS International Journal of Geo-Information |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2220-9964/10/7/485 |
| _version_ | 1797527025520476160 |
|---|---|
| author | Jiandong Bai Jiawei Zhu Yujiao Song Ling Zhao Zhixiang Hou Ronghua Du Haifeng Li |
| author_facet | Jiandong Bai Jiawei Zhu Yujiao Song Ling Zhao Zhixiang Hou Ronghua Du Haifeng Li |
| author_sort | Jiandong Bai |
| collection | DOAJ |
| description | Accurate real-time traffic forecasting is a core technological problem against the implementation of the intelligent transportation system. However, it remains challenging considering the complex spatial and temporal dependencies among traffic flows. In the spatial dimension, due to the connectivity of the road network, the traffic flows between linked roads are closely related. In the temporal dimension, although there exists a tendency among adjacent time points, the importance of distant time points is not necessarily less than that of recent ones, since traffic flows are also affected by external factors. In this study, an attention temporal graph convolutional network (A3T-GCN) was proposed to simultaneously capture global temporal dynamics and spatial correlations in traffic flows. The A3T-GCN model learns the short-term trend by using the gated recurrent units and learns the spatial dependence based on the topology of the road network through the graph convolutional network. Moreover, the attention mechanism was introduced to adjust the importance of different time points and assemble global temporal information to improve prediction accuracy. Experimental results in real-world datasets demonstrate the effectiveness and robustness of the proposed A3T-GCN. We observe the improvements in RMSE of 2.51–46.15% and 2.45–49.32% over baselines for the SZ-taxi and Los-loop, respectively. Meanwhile, the Accuracies are 0.95–89.91% and 0.26–10.37% higher than the baselines for the SZ-taxi and Los-loop, respectively. |
| first_indexed | 2024-03-10T09:38:04Z |
| format | Article |
| id | doaj.art-42a1d0d2797a41f2b0ee95de92e21b9c |
| institution | Directory Open Access Journal |
| issn | 2220-9964 |
| language | English |
| last_indexed | 2024-03-10T09:38:04Z |
| publishDate | 2021-07-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | ISPRS International Journal of Geo-Information |
| spelling | doaj.art-42a1d0d2797a41f2b0ee95de92e21b9c2023-11-22T03:55:14ZengMDPI AGISPRS International Journal of Geo-Information2220-99642021-07-0110748510.3390/ijgi10070485A3T-GCN: Attention Temporal Graph Convolutional Network for Traffic ForecastingJiandong Bai0Jiawei Zhu1Yujiao Song2Ling Zhao3Zhixiang Hou4Ronghua Du5Haifeng Li6Beijing Institute of Tracking and Telecommunication Technology, Beijing 100094, ChinaSchool of Geosciences and Info-Physics, Central South University, Changsha 410083, ChinaHuawei Technologies Co., Ltd., Shenzhen 518129, ChinaSchool of Geosciences and Info-Physics, Central South University, Changsha 410083, ChinaCollege of Automotive and Mechanical Engineering, Changsha University of Science and Technology, Changsha 410114, ChinaCollege of Automotive and Mechanical Engineering, Changsha University of Science and Technology, Changsha 410114, ChinaSchool of Geosciences and Info-Physics, Central South University, Changsha 410083, ChinaAccurate real-time traffic forecasting is a core technological problem against the implementation of the intelligent transportation system. However, it remains challenging considering the complex spatial and temporal dependencies among traffic flows. In the spatial dimension, due to the connectivity of the road network, the traffic flows between linked roads are closely related. In the temporal dimension, although there exists a tendency among adjacent time points, the importance of distant time points is not necessarily less than that of recent ones, since traffic flows are also affected by external factors. In this study, an attention temporal graph convolutional network (A3T-GCN) was proposed to simultaneously capture global temporal dynamics and spatial correlations in traffic flows. The A3T-GCN model learns the short-term trend by using the gated recurrent units and learns the spatial dependence based on the topology of the road network through the graph convolutional network. Moreover, the attention mechanism was introduced to adjust the importance of different time points and assemble global temporal information to improve prediction accuracy. Experimental results in real-world datasets demonstrate the effectiveness and robustness of the proposed A3T-GCN. We observe the improvements in RMSE of 2.51–46.15% and 2.45–49.32% over baselines for the SZ-taxi and Los-loop, respectively. Meanwhile, the Accuracies are 0.95–89.91% and 0.26–10.37% higher than the baselines for the SZ-taxi and Los-loop, respectively.https://www.mdpi.com/2220-9964/10/7/485traffic forecastingattention temporal graph convolutional networkspatial dependencetemporal dependence |
| spellingShingle | Jiandong Bai Jiawei Zhu Yujiao Song Ling Zhao Zhixiang Hou Ronghua Du Haifeng Li A3T-GCN: Attention Temporal Graph Convolutional Network for Traffic Forecasting ISPRS International Journal of Geo-Information traffic forecasting attention temporal graph convolutional network spatial dependence temporal dependence |
| title | A3T-GCN: Attention Temporal Graph Convolutional Network for Traffic Forecasting |
| title_full | A3T-GCN: Attention Temporal Graph Convolutional Network for Traffic Forecasting |
| title_fullStr | A3T-GCN: Attention Temporal Graph Convolutional Network for Traffic Forecasting |
| title_full_unstemmed | A3T-GCN: Attention Temporal Graph Convolutional Network for Traffic Forecasting |
| title_short | A3T-GCN: Attention Temporal Graph Convolutional Network for Traffic Forecasting |
| title_sort | a3t gcn attention temporal graph convolutional network for traffic forecasting |
| topic | traffic forecasting attention temporal graph convolutional network spatial dependence temporal dependence |
| url | https://www.mdpi.com/2220-9964/10/7/485 |
| work_keys_str_mv | AT jiandongbai a3tgcnattentiontemporalgraphconvolutionalnetworkfortrafficforecasting AT jiaweizhu a3tgcnattentiontemporalgraphconvolutionalnetworkfortrafficforecasting AT yujiaosong a3tgcnattentiontemporalgraphconvolutionalnetworkfortrafficforecasting AT lingzhao a3tgcnattentiontemporalgraphconvolutionalnetworkfortrafficforecasting AT zhixianghou a3tgcnattentiontemporalgraphconvolutionalnetworkfortrafficforecasting AT ronghuadu a3tgcnattentiontemporalgraphconvolutionalnetworkfortrafficforecasting AT haifengli a3tgcnattentiontemporalgraphconvolutionalnetworkfortrafficforecasting |