Comparative Analysis of Road Accident Prediction Using Ensemble Learning Method

The number of vehicles increasing on the road in the recent years which leads to increase in the number of accidents. Accident prediction and prevention is the major challenge faced by the government / transport department. The Objective of this system is to develop a machine learning model for real-time accident forecasting by comparing supervised algorithms with mean value of voting classifier results. Recent technologies like automated traffic control signals and IOT based GPS Technology helps in preventing accidents on the road. The Machine learning algorithms has been implemented to predict the occurrence of accidents on the road. Ensemble learning method is one of the best method for accident forecasting and finding the best road selection. This system is proposed to compare ensemble learning algorithm with other algorithms like supervised machine learning algorithm such as logistic regression, decision tree, random forest, and support vector classifier, K nearest neighbor and Naive Bayes. Ensemble Learning produces better predict performance compared to a single model. In Ensemble learning technique, various models will be combined and the best prediction result will be found. The comparative analysis helps to prove that the ensemble learning algorithm provides high accuracy of results than other model. The voting classifier method in ensemble learning helps to do comparative analysis and there by forecast accident and to find the best road. Dataset of previous accident reports available in government website has been used as input data to find the best road and to predict the accidents.