A Boosting-Based Machine Learning Approach for Intrusion Detection in Vehicle Address Verification Systems
Siva Gayatri Venkata Naga, Datta Sai Ammanamanchi, Angela Raj Chadha, Kethamreddy Karthikeya Reddy, Udhayakumar Shanmugam · 2024
Autonomous vehicles (AVs) are susceptible to cyberattacks due to their reliance on vehicle-to-everything (V2X) communication. This research proposes a Gradient Boosting-based intrusion detection system to safeguard AV networks. By employing advanced machine learning techniques, including Adaboost, Decision Trees, and Random Forests, the system effectively identifies various cyberattacks. Evaluated on standard datasets, the proposed IDS achieves a remarkable accuracy of 99.49%, surpassing existing models. The combination of feature selection and ensemble learning enhances the system's detection rate while maintaining computational efficiency.