AMFormer-based framework for accident responsibility attribution: Interpretable analysis with traffic accident features.

Accurately determining responsibility in traffic accidents is crucial for ensuring fairness in law enforcement and optimizing responsibility standards. Traditional methods predominantly rely on subjective judgments, such as eyewitness testimonies and police investigations, which can introduce biases...

Full description

Saved in:
Bibliographic Details
Main Authors: Yahui Wang, Zhoushuo Liang, Yue He, Jiahao Wu, Pengfei Tian, Zhicheng Ling
Format: Article
Language:English
Published: Public Library of Science (PLoS) 2025-01-01
Series:PLoS ONE
Online Access:https://doi.org/10.1371/journal.pone.0329107
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849248157660086272
author Yahui Wang
Zhoushuo Liang
Yue He
Jiahao Wu
Pengfei Tian
Zhicheng Ling
author_facet Yahui Wang
Zhoushuo Liang
Yue He
Jiahao Wu
Pengfei Tian
Zhicheng Ling
author_sort Yahui Wang
collection DOAJ
description Accurately determining responsibility in traffic accidents is crucial for ensuring fairness in law enforcement and optimizing responsibility standards. Traditional methods predominantly rely on subjective judgments, such as eyewitness testimonies and police investigations, which can introduce biases and lack objectivity. To address these limitations, we propose the AMFormer(Arithmetic Feature Interaction Transformer) framework-a deep learning model designed for robust and interpretable traffic accident responsibility prediction. By capturing complex interactions among key factors through spatiotemporal feature modeling, this framework facilitates precise multi-label classification of accident responsibility. Furthermore, we employ SHAP (SHapley Additive Interpretation) analysis to improve transparency by identifying the most influential features in attribution of responsibility, and provide an in-depth analysis of key features and how they combine to significantly influence attribution of responsibility. Experiments conducted on real-world datasets demonstrate that AMFormer outperforms both other deep learning models and traditional approaches, achieving an accuracy of 93.46% and an F1-Score of 93%. This framework not only enhances the credibility of traffic accident responsibility attribution but also establishes a foundation for future research into autonomous vehicle responsibility.
format Article
id doaj-art-1e97bec4d9124a22ba9922bedf62f77c
institution Kabale University
issn 1932-6203
language English
publishDate 2025-01-01
publisher Public Library of Science (PLoS)
record_format Article
series PLoS ONE
spelling doaj-art-1e97bec4d9124a22ba9922bedf62f77c2025-08-20T03:57:59ZengPublic Library of Science (PLoS)PLoS ONE1932-62032025-01-01207e032910710.1371/journal.pone.0329107AMFormer-based framework for accident responsibility attribution: Interpretable analysis with traffic accident features.Yahui WangZhoushuo LiangYue HeJiahao WuPengfei TianZhicheng LingAccurately determining responsibility in traffic accidents is crucial for ensuring fairness in law enforcement and optimizing responsibility standards. Traditional methods predominantly rely on subjective judgments, such as eyewitness testimonies and police investigations, which can introduce biases and lack objectivity. To address these limitations, we propose the AMFormer(Arithmetic Feature Interaction Transformer) framework-a deep learning model designed for robust and interpretable traffic accident responsibility prediction. By capturing complex interactions among key factors through spatiotemporal feature modeling, this framework facilitates precise multi-label classification of accident responsibility. Furthermore, we employ SHAP (SHapley Additive Interpretation) analysis to improve transparency by identifying the most influential features in attribution of responsibility, and provide an in-depth analysis of key features and how they combine to significantly influence attribution of responsibility. Experiments conducted on real-world datasets demonstrate that AMFormer outperforms both other deep learning models and traditional approaches, achieving an accuracy of 93.46% and an F1-Score of 93%. This framework not only enhances the credibility of traffic accident responsibility attribution but also establishes a foundation for future research into autonomous vehicle responsibility.https://doi.org/10.1371/journal.pone.0329107
spellingShingle Yahui Wang
Zhoushuo Liang
Yue He
Jiahao Wu
Pengfei Tian
Zhicheng Ling
AMFormer-based framework for accident responsibility attribution: Interpretable analysis with traffic accident features.
PLoS ONE
title AMFormer-based framework for accident responsibility attribution: Interpretable analysis with traffic accident features.
title_full AMFormer-based framework for accident responsibility attribution: Interpretable analysis with traffic accident features.
title_fullStr AMFormer-based framework for accident responsibility attribution: Interpretable analysis with traffic accident features.
title_full_unstemmed AMFormer-based framework for accident responsibility attribution: Interpretable analysis with traffic accident features.
title_short AMFormer-based framework for accident responsibility attribution: Interpretable analysis with traffic accident features.
title_sort amformer based framework for accident responsibility attribution interpretable analysis with traffic accident features
url https://doi.org/10.1371/journal.pone.0329107
work_keys_str_mv AT yahuiwang amformerbasedframeworkforaccidentresponsibilityattributioninterpretableanalysiswithtrafficaccidentfeatures
AT zhoushuoliang amformerbasedframeworkforaccidentresponsibilityattributioninterpretableanalysiswithtrafficaccidentfeatures
AT yuehe amformerbasedframeworkforaccidentresponsibilityattributioninterpretableanalysiswithtrafficaccidentfeatures
AT jiahaowu amformerbasedframeworkforaccidentresponsibilityattributioninterpretableanalysiswithtrafficaccidentfeatures
AT pengfeitian amformerbasedframeworkforaccidentresponsibilityattributioninterpretableanalysiswithtrafficaccidentfeatures
AT zhichengling amformerbasedframeworkforaccidentresponsibilityattributioninterpretableanalysiswithtrafficaccidentfeatures