Using Machine Learning Models to Forecast Severity Level of Traffic Crashes by R Studio and ArcGIS

This study describes crash causes, conditions, and distribution of accident hot spots along with an analysis the risk factors that significantly affect severity levels crashes their effects on pedestrian safety using machine learning (ML) techniques. Supervised ML algorithm–random forest decision tree–based algorithm-AdaBoost algorithms are applied compared to predict level future based road elements. Association rule, unsupervised algorithm, is utilized understand association between driver characteristics, geometric elements highway, environment, time, weather, speed. Slight, medium, severe injuries fatalities in also considered behavior drivers, who most likely cause crashes. Fatalities studied spatial statistics analysis. The variables affecting determined discussed detail. results checked for accuracy, sensitivity, specificity, recall, precision, F1 score performance. impact vehicles, characteristics investigated traffic random model was found be suitable algorithm levels.