Improving Detection and Prediction of Traffic Congestion in VANETs: An Examination of Machine Learning

Abstract

Traffic congestion remains a pressing challenge in urban areas, causing significant economic and environmental repercussions. To address this issue, accurate detection and prediction of traffic congestion are imperative for effective traffic management and planning. This research study investigates the efficacy of Support Vector Machines (SVM) and various other machine learning algorithms in augmenting traffic congestion detection and prediction for Vehicular Ad hoc Networks (VANETs). Leveraging historical congestion patterns, we train and evaluate the performance of the algorithms. Our results demonstrate the potential of SVM, coupled with advanced feature engineering techniques, to outperform other methods in accurately identifying and forecasting traffic congestion. The SVM classifier achieved an impressive classification accuracy of 0.99, showcasing its effectiveness in handling diverse traffic scenarios. Additionally, the K-Nearest Neighbors (KNN) and Ensemble Learning classifiers also yielded commendable accuracies of 0.99. Notably, the Decision Tree (DT) classifier attained a perfect accuracy score of 1.00, indicating its robustness in handling congestion patterns. The proposed approach not only achieves high detection accuracy but also exhibits remarkable robustness and scalability, enabling its application across various traffic scenarios. These findings contribute significantly to the development of intelligent traffic management systems, providing valuable insights into optimizing transportation networks. Ultimately, implementing our approach holds the potential to alleviate congestion, enhance travel efficiency, and foster urban sustainability.