Exploring Optimization and Machine Learning for Effective Intrusion Detection Systems
TL;DR
A comprehensive analysis of recent research investigates progress in IDS, highlighting the utilisation of Genetic Algorithms, Particle Swarm Optimisation, and ML models like Decision Trees, Support Vector Machines, and Neural Networks.
Abstract
Intrusion Detection Systems (IDS) are essential elements of contemporary cyber security frameworks, intended to identify unauthorised access and nefarious activity within networks and computer systems. This survey examines the classification of IDS technologies, methodologies, and approaches, emphasising the benefits and obstacles related to their use. Numerous optimisation methods in Intrusion Detection Systems (IDS) are examined, emphasising feature selection and machine learning (ML) algorithms that improve detection precision and efficacy. A comprehensive analysis of recent research investigates progress in IDS, highlighting the utilisation of Genetic Algorithms (GA), Particle Swarm Optimisation (PSO), and ML models like Decision Trees (DT), Support Vector Machines (SVR) and Neural Networks (NN). Furthermore, the emerging security concerns in Internet of Things (IoT) contexts and the significance of feature optimisation for enhancing IDS performance are discussed. This review synthesises ideas from current research to offer a thorough overview of IDS technologies and their function in enhancing network security against growing cyber threats.