.png){kind=icon}
.png){kind=icon}
Return
Representation learning of crash narrative using natural language processing models
.png){kind=icon}
PRE
AI
L
F
DOI：10.1016/j.jsr.2026.05.011.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
• Construct a continuous crash scenario space from narrative text using language models. • Embed and cluster crash narratives to reveal population-level crash structures. • Identify rare and atypical crash scenarios using Gaussian density estimation. • Introduce a visualization tool that supports navigating crash space.
Keywords:
crash scenario
natural language processing
representation learning
clustering
visualization
Journal
IF:
4.4
Papers:
2.7K
Citations:
6.7K
Organization
Cited Papers