An Ensemble-Based Machine Learning Model for Forecasting Network Traffic in VANET

Vehicular Ad-hoc Networks (VANETs), as the most significant element of Intelligent Transportation Systems (ITS), have potential to enhance traffic efficiency and road safety by making transportation system smarter are still at initial point development. In this paper, we propose an ensemble-based machine learning model for network prediction in VANET. We take advantage Ensemble Learning (EL), which combines different Machine (ML) models achieve better performance improve accuracy. consider informative attributes VANET dataset using Boruta LightGBM ensemble feature selection methods. Our proposed is based on Stacking with Booster Model (STK–EBM) designed a stacking heterogeneous ML models. The framework consists two layers, including base layer meta layer. first integrates Random Forest (RF), K-Nearest Neighbor (KNN) XGBoost booster learners. An optimized Logistic Regression (LR) employs our learner second evaluate considering classification metrics then compare it popular predictive Simulation results show that STK–EBM gives more stable than single algorithm, well overall terms accuracy execution time.