A Comparative Analysis of Network Intrusion Detection Systems in Machine Learning and Deep Learning for Securing IoT devices
Maganti Sri Ramya, C. Santhanakrishnan, M Janani, S. Poonkodi · Advances in computational intelligence and robotics book series · 2025
Now a day we can see that the internet is booming because of that there is growing number of users for the internet usage. Basically, these users are creating a lot of traffic while connecting to any site or during any using their services. Intrusion Detection Systems (IDS) play an important role in defending computer networks from various cyber threats. The usefulness of Machine Learning (ML) and Deep Learning (DL) models in IDS has grown in importance in tackling network security challenges in recent years. The study examines the performance of traditional ML algorithms, alongside advanced DL models. Such insights are essential for enhancing network security in an ever-evolving threat landscape. The results of this study can guide security practitioners and system administrators in selecting the most suitable model for their specific network security needs. The chapter provides an overview of prior work on ML and DL in IDS. In addition, ML and DL models were examined, and results were presented, contrasted, and discussed using two dataset NSL-KDD Dataset, KDD99.