Artificial Intelligence based Intrusion Detection Systems using Deep Learning Techniques
Alycia Sebastian, Hamed Al Hajri · 2024
Security in the cyber world is a critical component of the field of cyber security, and it involves managing cyber threats and risks. An Intrusion Detection System (IDS) scans a network for vulnerabilities and so that the users or business organizations can take measures against any breaches that are discovered. An IDS defends the computer's infrastructure against unauthorized access by users, including the insiders. Deep Learning (DL) is one of the intriguing approaches that are currently utilized by IDS to improve their effectiveness in protecting networks of computers and servers. In supervised learning approaches classifying big data is a challenge and it requires a labelled dataset. The study explores and examines the unsupervised DL methods, AutoEncoder(AE) and Deep Belief Networks (DBN) for detecting the abnormal behaviours in the networks and suitability of these methods for IDS.