Identify Abnormal System Behaviors to Improve Security in Evolutionary Algorithm-Based Cloud Computing

Abstract

Information security in virtual environments and new areas called cloud computing have always been emphasized as one of the basic infrastructures and requirements in the developmental and inclusive use of ICT. Although absolute security is unattainable both in the real environment and in the cyberspace, it is possible to create a level of security that is sufficient and appropriate in almost all environmental conditions. There are many security challenges in cloud computing that need to be addressed by cloud service providers to persuade users to use the technology. One of the most important issues is ensuring user data is accurate and inaccessible. For the user, the security process used to store data in the cloud is very vague, lengthy, and unclear. In this study, an abnormal behavior-based security approach is designed to detect events that appear to be abnormal compared to other normal system behaviors. The focus of this paper is on the use of evolutionary algorithms such as genetic algorithm (GA) or other new algorithms such as colonial competition algorithm (CCA) to detect these abnormal behaviors with the help of intelligent learning agents. Similar studies have used different optimization methods such as PSO algorithm and so on. The proposed algorithms can be compared and evaluated with previous methods.