Traffic Density Classification Using Multilayer Perceptron and Random Forest Method

Traffic management is done to overcome congestion due to overcrowding and overcapacity. However, this arrangement still utilizes information obtained from various entities on the road, namely the police and transportation service officers. Observation of conditions and situations on the road is still subjective, so traffic management becomes subjective. However, there are potential technologies that can be utilized to help the existing problems. With these problems and opportunities, there is in providing traffic density information that is more objective utilizing the latest technology. The development of various types of information system adaptation and the use of technology is able to provide information on a regular basis. Machine learning as a form of technology development that is being optimized, can solve the information needs typical of traffic control. In this study, a traffic density classification model was made using an algorithm based on Artificial Neural Network-Multilayer Perceptron and Random Forest. The application of this research is carried out in five stages, namely understanding business needs, understanding data, cleaning and preparing data, optimizing parameters and modeling, and evaluating. By, using the method, Artificial Neural Network gives the optimum result and can help traffic management system.