Comparative Analysis of RF, Naive Bayes and GRU Algorithms for Network Traffic Prediction in VANET

Ensuring the Vehicular Ad-Hoc Networks (VANETs) are the critical component of Intelligent Transportation Systems (ITS), which enables vehicles to communicate and exchange data in real- time. Accurate network traffic prediction in VANETs is essential to ensure efficient communication, reduce congestion, and improve road safety. In this research paper, we propose a hybrid model combining of Random Forest (RF) and Naive Bayes (NB) algorithms for prediction of network traffic in VANET environments. The RF algorithm captures complex, non-linear relationships in network traffic data, while the NB classifier leverages probabilistic reasoning to handle uncertainties in traffic patterns. Our proposed model is evaluated on real-world VANET datasets, focusing on parameters such as vehicle density, speed, and communication range. The results demonstrate that the RF-NB hybrid approach improves prediction accuracy compared to standalone models. (with parameters and its values) This prediction framework can significantly enhance V ANET performance by enabling proactive traffic management and optimized data routing, thereby contributing to safer and more efficient vehicular communication systems.