Urban Road Accident Evaluation And Road Accident Severity Prediction

  • Anant Ram, B.Sikander


Road accidents have emerged into a world issue .Based on the latest World Health Organization (WHO) report, road accidents are the world's 10th major cause of death. This has become a major problem in many developed and developing countries due to the large number of road accidents each year. It is completely inadmissible and saddening to enable road accidents to destroy the residents. Therefore, a detailed analysis is needed to handle this overwhelmed situation. In this paper, the research study seeks to use machine-learning models to more closely investigate road accidents and to estimate the rate of accidents in India. In this article, we also describe some factors that specifically influence road accidents and provide some useful suggestions on the matter. Here we considered five popular classification methods to build an accurate prediction model, such as Naive Bayes, Logistic Regression, Decision tree. The result is given by these methods is somewhat the same so then we used the ensemble learning concept i.e. Random Forest, lastly, we used a support vector machine to classify road accidents into Fatal, serious & Slight accidents. Ultimately, Random Forest provides the best and most effective result.

How to Cite
Anant Ram, B.Sikander. (2020). Urban Road Accident Evaluation And Road Accident Severity Prediction. International Journal of Advanced Science and Technology, 29(08), 4086-4097. Retrieved from https://sersc.org/journals/index.php/IJAST/article/view/26224