Meta-heuristics based Intrusion Detection System: A Comparative Analysis

Abstract

Cloud computing uses a pay-per-use model to provide a variety of services over the Internet. As a result, many corporations have already implemented this framework to entice consumers with its appealing features. It is, however, vulnerable to malware due to its architecture. It necessitates the use of an Intrusion Detection System (IDS) capable of detecting such attacks in a cloud environment with a high accuracy rate. The analysis of IDS has received a great deal of emphasis due to the growing extent of network throughput and security threats. Meta-Heuristics (MH) utilizes the concept based on the natural computing principles for high level optimization solutions. MH began to attract the attention of researchers employed in intrusion detection system (IDS) after being successfully applied in other fields. This paper investigates the role of existing MH algorithms for the feature optimization problem in IDS. The quantitative analysis of five natural computing algorithms such as, Genetic Algorithm (GA), Particle Swarm Optimization (PSO), Cuckoo Search (CS), Ant Colony Optimization (ACO), and Artificial Bee Colony (ABC) is performed in the paper. It motivates the researchers for the integration of various meta-heuristic approaches and provides a clear idea about which approach is best suitable for what type of optimization problem or situation.