Predicting multiple types of traffic accident severity with explanations: A multi-task deep learning framework
Predicting traffic accident severity is essential for traffic accident prevention and vulnerable road user safety. Furthermore, the explainability of the prediction is crucial for practitioners to extract relevant risk factors and implement corresponding countermeasures. Most extant research ignores the property loss severity of traffic accidents and fails to predict different levels of death and property loss severity. Moreover, while the explainability of traditional models is easy to achieve, an explainable design of deep neural network (DNN) is extremely deficient in existing research. Few attempts that incorporate neural networks suffer from the lack of multiple hidden layers and the negligence of structural information when explaining predictions. In this study, the authors propose a multi-task DNN framework for predicting different levels of injury, death, and property loss severity. The multi-task and deep learning design enables a comprehensive and precise analysis of traffic accident severity. Unlike many black-box DNN algorithms, the framework could identify key factors that cause the three types of traffic accident severity via layer-wise relevance propagation, which generates explanations based on the structure and weights of DNN. Based on the experiments conducted using Chinese traffic accident data, the proposed model predicts traffic accident severity risks with good accuracy and outperforms state-of-the-art methods. Furthermore, the case studies show that the key factors provided by the framework are more reasonable and informative than the explanations provided by baseline methods. The model is the first multi-task learning model and the first DNN-based model for traffic accident severity prediction to the best of the authors’ knowledge.
- Record URL:
- Record URL:
-
Availability:
- Find a library where document is available. Order URL: http://worldcat.org/issn/09257535
-
Supplemental Notes:
- © 2021 Elsevier Ltd. All rights reserved. Abstract reprinted with permission of Elsevier.
-
Authors:
- Yang, Zekun
-
0000-0003-4040-8476
- Zhang, Wenping
-
0000-0002-0183-4504
- Feng, Juan
-
0000-0002-0548-1531
- Publication Date: 2022-2
Language
- English
Media Info
- Media Type: Web
- Features: References;
-
Serial:
- Safety Science
- Volume: 146
- Issue Number: 0
- Publisher: Elsevier
- ISSN: 0925-7535
- Serial URL: http://www.sciencedirect.com/science/journal/09257535
Subject/Index Terms
- TRT Terms: Crash injuries; Crash severity; Mathematical prediction; Neural networks
- Geographic Terms: China
- Subject Areas: Data and Information Technology; Highways; Safety and Human Factors;
Filing Info
- Accession Number: 01786123
- Record Type: Publication
- Files: TRIS
- Created Date: Oct 26 2021 2:26PM