Using data analytic and machine learning for road traffic accidents prediction and causal factors analysis

Abstract

Machine learning and data analysis together represent a powerful force in producing analysis and forecasting models which in turn support the growth of different fields. In terms of safety and people, road traffic accidents have a negative impact on individuals, governments, and companies as well. Therefore, predicting traffic accidents on the road has become very important in the decision-making process that will lead to the safety of others. It is also important to find the causes of these accidents. This study aims to predict road traffic accidents and their causative factors based on a data set from Kaggle [1]. And the establishment of a vehicle accident prediction model on the road using machine learning and a decision tree algorithm, and the results showed that one of the most important factors causing accidents on the road is the lack of distance between vehicles, and that weather and age play an important role, and the model achieved an accuracy of 84.22% and a standard deviation of 1.08 %+/-. The tools used in data analysis and machine learning and prediction are Python and RapidMiner an integrated software platform for data science that provides an integrated environment for machine learning, deep learning, and predictive analysis.