Détection d'intrusions réseaux en utilisant le Machine Learning

Abstract

We are currently witnessing a constant increase in threats related to network intrusions. In this study, we focus on network intrusion detection using Machine Learning. Our objective is to explore different detection techniques by leveraging Machine Learning models. The results we obtain are carefully evaluated and compared to determine the most effective models. We highlight the importance of Machine Learning in intrusion detection, as it enables early and accurate detection of network intrusions. Additionally, we identify opportunities for future research to enhance intrusion prevention and response techniques, ultimately ensuring enhanced network security. In conclusion, XGBoost is recommended for network intrusion detection due to its high accuracy, ability to minimize classification errors, and good performance in terms of recall.