Temporal Analysis Based-Machine Learning of Road Traffic Accident Fatality Prediction

View Full Paper

Ask AI

Bookmark

Random Forest models achieved an accuracy ranging from 96.65% to 97.25%, demonstrating high predictive power.
The research utilized over 5.4 million accident and vehicle records from UK datasets spanning nearly 30 years.
Stratified 5-fold cross-validation and Synthetic Minority Oversampling Technique aided in developing robust classification models.
Identified risk factors like urban or rural areas and speed limits remained stable over three decades, informing preventive measures.

Discussion

No takes yet. Share an insight, caveat, or question.

Observational analysis improves prediction of RTA fatalities using machine learning methods, highlighting risk factors.

Lee et al. (2025) studied this question.

View Full Paper

Ask AI

Bookmark