Prediction of Cyclist Road Accidents in India using Machine Learning and Visualization Techniques

Abstract

Road Traffic Accidents (RTAs) are a major problem in today's world, as their number continues to grow every year due to an increase in the use of road transport and a variety of other factors. This paper describes the best algorithm model to be used to improve the accuracy and predictability of cyclist accidents. All datasets are imported from traffic incidents, taken from the Indian dataset STATS2021 for the year 2020. The knowledge derived from these large datasets forms the basis for a variety of predictions. The handling of an imbalanced dataset are performed by data mining methods, such as, Naive Bayes, Logistic Regression, Random Forest and Decision Tree, which are also used for the prediction purpose. The project defines the most suitable classification algorithm that can be used to predict cyclist casualties in the future. In addition, the key factors causing these road accidents are identified and presented.