- Title
- TAP : Traffic Accident Profiling via multi-task spatio-temporal graph representation learning
- Creator
- Liu, Zhi; Chen, Yang; Xia, Feng; Bian, Jixin; Zhu, Bing; Shen, Guojiang; Kong, Xiangjie
- Date
- 2023
- Type
- Text; Journal article
- Identifier
- http://researchonline.federation.edu.au/vital/access/HandleResolver/1959.17/194437
- Identifier
- vital:18358
- Identifier
-
https://doi.org/10.1145/3564594
- Identifier
- ISSN:1556-4681 (ISSN)
- Abstract
- Predicting traffic accidents can help traffic management departments respond to sudden traffic situations promptly, improve drivers' vigilance, and reduce losses caused by traffic accidents. However, the causality of traffic accidents is complex and difficult to analyze. Most existing traffic accident prediction methods do not consider the dynamic spatio-temporal correlation of traffic data, which leads to unsatisfactory prediction accuracy. To address this issue, we propose a multi-task learning framework (TAP) based on the Spatio-temporal Variational Graph Auto-Encoders (ST-VGAE) for traffic accident profiling. We firstly capture the dynamic spatio-temporal correlation of traffic conditions through a spatio-temporal graph convolutional encoder and embed it as a low-latitude vector. Then, we use a multi-task learning scheme to combine external factors to generate the traffic accident profiling. Furthermore, we propose a traffic accident profiling application framework based on edge computing. This method increases the speed of calculation by offloading the calculation of traffic accident profiling to edge nodes. Finally, the experimental results on real datasets demonstrate that TAP outperforms other state-of-the-art baselines. © 2023 Copyright held by the owner/author(s). Publication rights licensed to ACM.
- Publisher
- Association for Computing Machinery
- Relation
- ACM Transactions on Knowledge Discovery from Data Vol. 17, no. 4 (2023), p.
- Rights
- All metadata describing materials held in, or linked to, the repository is freely available under a CC0 licence
- Rights
- © 2023 Copyright held by the owner/author(s)
- Subject
- 4606 Distributed computing and systems software; 4604 Cybersecurity and privacy; Graph convolutional network; Graph Representation learning; Spatio-temporal data; Traffic accident profiling
- Reviewed
- Funder
- This work is supported by the National Natural Science Foundation of China (62072409 and 62073295), Zhejiang Province Basic Public Welfare Research Project (LGG20F030008), Zhejiang Provincial Natural Science Foundation (LR21F020003), and Fundamental Research Funds for the Provincial Universities of Zhejiang (RF-B2020001).
- Hits: 328
- Visitors: 303
- Downloads: 0
| Thumbnail | File | Description | Size | Format |
|---|