The Key Determinants of Road Accidents: A Machine Learning Analysis
摘要
The significance of road safety increases with the increase in traffic volume and rapid urbanization. Despite their significant importance, there is a lack of research on road accidents on this major transportation corridor in Bangladesh. This paper deals with the analysis of accident data from Pabna-Ishwardi, Ruppur-Dashuria Highway, a very important route to support logistics for the Ruppur Nuclear Power Plant. This study preprocesses the dataset with various encoding techniques and uses PyCaret to classify the target column, “Cause of Accident”, identifying key patterns in road accidents. Ridge Classifier, which was chosen and evaluated using PyCaret, gave the best result with an accuracy of 76.58%, AUC of 0.8135, F1 score of 0.6954, and recall 0.7658. The model identifies “Time” and “Vehicular Involvement” as major causes of collisions and gives practical advice that could reduce collision rates and increase safety both for cars and pedestrians along this important corridor.