A SURVEY ON NETWORK-BASED INTRUSION DETECTION SYSTEMS USING MACHINE LEARNING ALGORITHMS

Abstract

Network security is of central significance in the current information world. Due to the rapid increase of network-enabled devices, there is a significant risk of network intrusion more than ever. Hackers and intruders can successfully attack to cause the crash of the networks and web services by the unauthorized intrusion, which may cause a significant loss to an organization in terms of data and money. So, it is high time to create an intrusion detection system that can detect all types of intrusion. Due to the rapid growth and significant results of machine learning (ML) algorithms in several areas, there has recently been much interest in applying them to network security. The network-based intrusion detection system (NIDS) has much promise to be the borderline of defense against intrusions in the current information communication technology (ICT) era, and it's a critical aspect of network security. Due to the dynamic nature of attacks, intrusion detection datasets are available publicly. Intrusion detection systems are the backbone of the networks and data protection. Various IDS approaches have been used over time to achieve maximum detection accuracy. This paper investigates the different machine learning methods used to deploy network-based intrusion detection systems. This survey could give scholars a better grasp of present methodologies and help them find research possibilities and do further research in this area.