Using improved neural network for the risk assessment of information security

Abstract

Abstract Because of the information age, protecting information is very important to satisfy the three main aspects of information security, namely confidentiality, integrity and availability. In this case, information security has become one of the most important problems in information technology. Information security is a very important activity and risk assessment is the kernel of information security. However, most of the current risk assessment activities are comparatively subjective and the performances are not good enough. To understand this problem, we propose the improved neural network for the risk assessment of information security. Basically, it is processed under back-propagation neural network (BPN). Moreover, particle swarm optimization (PSO) is used for fine parameter optimization of BPN. The experimental results show that the proposed algorithm has the best performance among these compared approaches.