Traffic Accidents Analysis with respect to Road Users using Data mining Techniques

Abstract: Analysis of road accidents is a very significant because it can expose the association between the diverse types of attributes that contributes to a road accident. Attributes that affect the road accidents can be road attribute, environment attributes, traffic attributes etc. Analyzing road accidents can provide the information about the role of these attributes which can be utilized to defeat the accident rate. There is an enormous range of research work that has already contributed to the road accident analysis. Few studies concentrated on determining the related state of affairs behind the occurrence of road accident as well as few studies concentrated on determining the related circumstances with the brutality of the accident. In these days, Data mining has often utilized a technique for probing the dataset of a road accident. In this study, we applied the classification analysis of data and we achieved some results and later we implemented different clustering techniques such as Self Organizing map (SOM), K-modes and classifications techniques such as Support vector machine (SVM), Naive Bayes (NB) and Decision tree to achieve better accuracy on the basis of victim class and we achieved better results by using SOM instead of using k-modes.