A review on recent approaches of Machine Learning, Deep Learning, and Explainable Artificial intelligence in Intrusion Detection Systems

In recent decades, network security has become increasingly crucial, and intrusion detection systems play a critical role in securing it. An intrusion Detection System (IDS) is a mechanism that protects the network from various possible intrusions by analyzing network traffic. It provides confidentiality and ensures the integrity and availability of data. Intrusion detection is a classification task that classifies network data into benign and attack by using various machine learning and deep learning models. It further develops a better potential solution for detecting intrusions across the network and mitigating the false alarm rate efficiently. This paper presents an overview of current machine learning (ML), deep learning (DL), and eXplainable Artificial intelligence (XAI) techniques. Our findings provide helpful advice to researchers who are thinking about integrating ML and DL models into network intrusion detection. At the conclusion of this work, we outline various open challenges.

Keywords:

Network Security , IPS , Preprocessing , SMOTE , Datasets , Attacks , Feature Selection