Intelligent transportation system for internet of vehicles based vehicular networks for smart cities

Abstract

A popular research area is internet traffic analysis as it has many applications, mainly for classifying internet traffic. Innovative technologies have been developed for predicting and identifying traffic congestion in the intelligent Internet of vehicles (IOVs). In this paper, an intelligent transport system for the IOVs-based vehicular network traffic for smart city scenario is proposed based on tree-based Decision Tree (DT), Random Forest (RF), and Extra Tree (ET), and XGBoost machine learning (ML) models. Simulation results indicate that the proposed system can provide high detection accuracy and low computational costs thanks to ensemble learning and averaging important feature selection. The tree-based ML techniques with feature selection performed better than those without feature selection for IOV-based vehicular network traffic. The Stacking model shows higher classification accuracy, 99.05%, compared to the lowest KNN accuracy, 96.6%, and SVM accuracy, 98.01%.