An Efficient Analysis of Road Accidents Severity Using Novel Support Vector Machines Over Artificial Neural Networks with Improved Accuracy Rate

Abstract

Aim: To perform an efficient analysis of road accidents severity prediction using Support Vector Machine and Artificial Neural Network algorithms. Materials and Methods: Road accident severity prediction is performed by Support Vector Machine algorithm (N=10) and Artificial Neural Network algorithm (N=10) using Traffic-crashes dataset. Results and Discussion: The road accident severity prediction using Support Vector Machine algorithm (Accuracy 83.6%) and with Artificial Neural Network algorithm (Accuracy 73.3%). There is no significant difference between the groups. Conclusion: Within the limits of this study, SVM algorithm has significantly better accuracy than ANN algorithm.