Simulation of Drivers’ Behavior in Traffic Based on Data Mining

Due to the phenomenon of urbanization in Iran and the increasing number of cars in cities, we are witnessing an exponential growth of traffic in the city's streets, which has become a matter of concern in large cities. This paper studies different groups of drivers in a district and identifies characteristics features, practical methods, and models in changing their behavioral patterns. Simulation tools and layout of the correct scenarios are implemented to reduce traffic. To achieve the purpose of the research, data mining techniques were performed on traffic data in Karaj and Tehran cities. Considering the application of data mining algorithms on traffic data, observing the outputs, and finally, the extracted result from the combination of two clustering algorithms and the decision tree provided a model with the most acceptable answer. Studies show the accuracy of its function with a high percentage. We can predict dangerous behaviors, which significantly reduce traffic.