A Review on Challenges and Future Research Directions for Machine Learning-Based Intrusion Detection System

Research in the field of Intrusion Detection is focused on developing an efficient strategy that can identify network attacks. One of the important strategies is to supervise the network events for identifying attacks. Security mechanisms such as Intrusion Detection Systems (IDS) have been used for securing the network infrastructure and network communication against network attacks, wherein Machine Learning (ML) techniques have a notable contribution to design an efficient IDS. However, dependence on modern communication technology and collateral rise in the network attacks affect the performance of ML techniques. In this article, we discuss a detailed overview of intrusion detection using ML techniques. We discuss the steps performed by ML techniques for detecting and classifying intrusions. Moreover, our paper provides a comprehensive overview of state-of-the-art ML techniques used for intrusion detection and classification along with their advantages and limitations. The paper also summarizes research work performed in the field of ML-based IDS. In this paper, we aim to discuss various challenges faced by ML-based IDS. We further discuss future research directions that can be considered for enhancing the efficiency and effectiveness of IDS. Our review will serve as an incentive to novice researchers who aim to work in the field of ML-based IDS.