Measuring the Risk of Accidents Using K-Means Clustering Algorithm

Abstract

There are many automobile industries to design and build safety measures for automobiles, but it is difficult to predict traffic accidents. There are a large number of accidents happening in rural and urban areas. As the driver is the information receiver and prime decision-maker in the driving process, our research focuses to make the driver aware of the risks and consequences in the route that the driver wanted to take and proceed. Patterns associated with different attributes can be identified by developing an accurate prediction analysis which will be capable of automatic prediction of various accident scenarios. Our model will be useful to prevent accidents and identify areas and conditions in which accidents are likely to happen. That will further help in the development of safety measures. We are aiming to acquire maximum possibilities using previous records applying the k-means algorithm.