Spatial‐temporal attention wavenet: A deep learning framework for traffic prediction considering spatial‐temporal dependencies
Abstract Traffic prediction on road networks is highly challenging due to the complexity of traffic systems and is a crucial task in successful intelligent traffic system applications. Existing approaches mostly capture the static spatial dependency relying on the prior knowledge of the graph struct...
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Wiley
2021-04-01
|
| Series: | IET Intelligent Transport Systems |
| Subjects: | |
| Online Access: | https://doi.org/10.1049/itr2.12044 |
| _version_ | 1828237067299061760 |
|---|---|
| author | Chenyu Tian Wai Kin (Victor) Chan |
| author_facet | Chenyu Tian Wai Kin (Victor) Chan |
| author_sort | Chenyu Tian |
| collection | DOAJ |
| description | Abstract Traffic prediction on road networks is highly challenging due to the complexity of traffic systems and is a crucial task in successful intelligent traffic system applications. Existing approaches mostly capture the static spatial dependency relying on the prior knowledge of the graph structure. However, the spatial dependency can be dynamic, and sometimes the physical structure may not reflect the genuine relationship between roads. To better capture the complex spatial‐temporal dependencies and forecast traffic conditions on road networks, a multi‐step prediction model named Spatial‐Temporal Attention Wavenet (STAWnet) is proposed. Temporal convolution is applied to handle long time sequences, and the dynamic spatial dependencies between different nodes can be captured using the self‐attention network. Different from existing models, STAWnet does not need prior knowledge of the graph by developing a self‐learned node embedding. These components are integrated into an end‐to‐end framework. The experimental results on three public traffic prediction datasets (METR‐LA, PEMS‐BAY, and PEMS07) demonstrate effectiveness. In particular, in the 1 h ahead prediction, STAWnet outperforms state‐of‐the‐art methods with no prior knowledge of the network. |
| first_indexed | 2024-04-12T20:43:44Z |
| format | Article |
| id | doaj.art-4e5de810408b4e518e65468b443200b2 |
| institution | Directory Open Access Journal |
| issn | 1751-956X 1751-9578 |
| language | English |
| last_indexed | 2024-04-12T20:43:44Z |
| publishDate | 2021-04-01 |
| publisher | Wiley |
| record_format | Article |
| series | IET Intelligent Transport Systems |
| spelling | doaj.art-4e5de810408b4e518e65468b443200b22022-12-22T03:17:21ZengWileyIET Intelligent Transport Systems1751-956X1751-95782021-04-0115454956110.1049/itr2.12044Spatial‐temporal attention wavenet: A deep learning framework for traffic prediction considering spatial‐temporal dependenciesChenyu Tian0Wai Kin (Victor) Chan1Tsinghua‐Berkeley Shenzhen Institute Tsinghua Univeristy Shenzhen 510006 People's Republic of ChinaTsinghua‐Berkeley Shenzhen Institute Tsinghua Univeristy Shenzhen 510006 People's Republic of ChinaAbstract Traffic prediction on road networks is highly challenging due to the complexity of traffic systems and is a crucial task in successful intelligent traffic system applications. Existing approaches mostly capture the static spatial dependency relying on the prior knowledge of the graph structure. However, the spatial dependency can be dynamic, and sometimes the physical structure may not reflect the genuine relationship between roads. To better capture the complex spatial‐temporal dependencies and forecast traffic conditions on road networks, a multi‐step prediction model named Spatial‐Temporal Attention Wavenet (STAWnet) is proposed. Temporal convolution is applied to handle long time sequences, and the dynamic spatial dependencies between different nodes can be captured using the self‐attention network. Different from existing models, STAWnet does not need prior knowledge of the graph by developing a self‐learned node embedding. These components are integrated into an end‐to‐end framework. The experimental results on three public traffic prediction datasets (METR‐LA, PEMS‐BAY, and PEMS07) demonstrate effectiveness. In particular, in the 1 h ahead prediction, STAWnet outperforms state‐of‐the‐art methods with no prior knowledge of the network.https://doi.org/10.1049/itr2.12044Traffic engineering computingCombinatorial mathematicsNeural nets |
| spellingShingle | Chenyu Tian Wai Kin (Victor) Chan Spatial‐temporal attention wavenet: A deep learning framework for traffic prediction considering spatial‐temporal dependencies IET Intelligent Transport Systems Traffic engineering computing Combinatorial mathematics Neural nets |
| title | Spatial‐temporal attention wavenet: A deep learning framework for traffic prediction considering spatial‐temporal dependencies |
| title_full | Spatial‐temporal attention wavenet: A deep learning framework for traffic prediction considering spatial‐temporal dependencies |
| title_fullStr | Spatial‐temporal attention wavenet: A deep learning framework for traffic prediction considering spatial‐temporal dependencies |
| title_full_unstemmed | Spatial‐temporal attention wavenet: A deep learning framework for traffic prediction considering spatial‐temporal dependencies |
| title_short | Spatial‐temporal attention wavenet: A deep learning framework for traffic prediction considering spatial‐temporal dependencies |
| title_sort | spatial temporal attention wavenet a deep learning framework for traffic prediction considering spatial temporal dependencies |
| topic | Traffic engineering computing Combinatorial mathematics Neural nets |
| url | https://doi.org/10.1049/itr2.12044 |
| work_keys_str_mv | AT chenyutian spatialtemporalattentionwavenetadeeplearningframeworkfortrafficpredictionconsideringspatialtemporaldependencies AT waikinvictorchan spatialtemporalattentionwavenetadeeplearningframeworkfortrafficpredictionconsideringspatialtemporaldependencies |