Identification of high-risk roadway segments for wrong-way driving crash using rare event modeling and data augmentation techniques

•WWD crashes are relatively rare but more likely to produce fatalities and severe injuries than other types of crashes.•To solve the imbalanced dataset problem data augmentation and rare event modeling techniques were applied.•A hybrid data augmentation process was adopted in this study using SMOTE,...

Ausführliche Beschreibung

Bibliographische Detailangaben

Veröffentlicht in:	Accident analysis and prevention 2023-03, Vol.181, p.106933-106933, Article 106933
Hauptverfasser:	Ashraf, Md Tanvir, Dey, Kakan, Mishra, Sabyasachee
Format:	Artikel
Sprache:	eng
Schlagworte:	Accidents, Traffic Autoencoder Automobile Driving Data augmentation Deep Learning Florida Humans Rare event modeling Traffic safety Wrong-way
Online-Zugang:	Volltext
Tags:	Tag hinzufügen Keine Tags, Fügen Sie den ersten Tag hinzu!