Enhancing Intrusion Detection Systems using Ensemble Machine Learning Techniques

Abstract

The increasing usage of the Internet has also brought about the risk of network attacks, leading to the need for effective intrusion detection systems. This chapter aims to fill the gap in literature by conducting a comprehensive review of 55 relevant studies conducted from 2000 to 2007, focusing on the use of machine learning techniques for intrusion detection. The reviewed studies are compared based on the design of their classifiers, the datasets used in their experiments, and other experimental setups. Single, hybrid, and ensemble classifiers are examined, and their achievements and limitations are discussed. The chapter provides a thorough evaluation of the strengths and weaknesses of using machine learning for intrusion detection and suggests future research directions in this field. In conclusion, this chapter addresses the need for a comprehensive review of machine learning techniques in intrusion detection. It provides insights into classifier design, dataset selection Other experimental details an assessment of the use of machine learning for intrusion detection is presented, and recommendations for future studies are suggested.