Driver Behavior Pro ling using Machine Learning

Abstract

The drivers’ behavior influences the traffic on road and this, in turn influences energy consumed by the vehicles and emission of pollutants from the vehicles. So it is necessary to identify drivers’ characteristics to profile their behav-ior correctly. A large amount of data is needed for the analysis which is collected by the on-Board Unit present on the vehicle. On-Board Unit has sensors that are used to collect the required data. The comparative performance of different ma-chine learning algorithms is evaluated on the data collected by the on-board unit and in turn help in profiling drivers’ behavior. The experimental result shows that the support vector machine gives an accuracy of 99.4% amongst the remaining classifier.