Comparative Study of Machine Learning Algorithms using the CICIOV2024 Dataset

Abstract

Leveraging the CICIoV2024 dataset, this study evaluates the performance of different algorithms in accurately identifying and classifying intrusions in vehicular networks, specifically using LightGBM, XGBoost, CatBoost, and LCCDE algorithms. A comparative analysis conducted considered key performance metrics such as accuracy, precision, recall, and F1-score, shedding light on the strengths and limitations of each approach. Through rigorous experimentation and evaluation, the remarkable results demonstrated 100% accuracy, recall, precision, and F1-score across all models. These findings highlight the efficacy of employing advanced machine learning techniques for enhancing intrusion detection capabilities in IoV environments. The insights gained from this study can inform the development of highly accurate and reliable intrusion detection systems tailored to the unique challenges of vehicular networks.Keywords: CatBoost, CICIoV2024, IoV, LCCDE, LightGBM, XGBoost