Using Multidimensional Analysis to Predict the Severity of Traffic Accidents and Their Contributing Factors

School of Economics and Management, Beijing Jiaotong University (Weihai), 264400 Weihai City, Shandong Province, China

^* Corresponding author’s email: This email address is being protected from spambots. You need JavaScript enabled to view it.

Abstract

In order to forecast the severity of traffic accidents and investigate the factors that influence their severity, the study uses data on UK traffic accidents from the Kaggle website. Three machine learning models—Random Forest, SVM and CatBoost—are used in this work, and their predictive powers are compared. These models’ forecast accuracy for serious and fatal traffic accidents was lower, although their overall good accuracy. The prediction results demonstrated that the severity of traffic accidents is significantly influenced by a number of parameters, including the time of day, the road type, the speed-limit, and weekends. Road safety hazards can be identified and managed with the use of road accident severity prediction. In order to improve prediction accuracy, future studies should include more data dimensions and use models or algorithms that are more suited to managing unbalanced data. Richer insights for managing traffic safety and developing policies would result from this.