A Novel Computer Virus Detection Method from Ideas of Immunology

Abstract

In this paper, inspired by the change of antibody concentration in biological immune system, we present a Novel Computer Virus Detection Method from Ideas of Immunology (NCVDMII). In NCVDMII, the concepts and formal definitions of antibody, antigen, nonself and detector are given. Then, NCVDMII get the specific antibody from the simulation of primary immune response and calculate the antibody number by using the secondary immune response. Finally, building upon the above processes, a probabilistic calculation method for computer virus alarm production, which is based on the correlation between the antibody concentration and the computer virus infection intensity, is proposed. Compared with Negative Selection Algorithm (NSA), Automatic Computer Immune System (ACIS) and Immunity-based System (IBS), the false alarm rate of NCVDMII is reduced by 17.5%, 5.0% and 2.8%, respectively.

Research on Network Malicious Code Immune Based on Imbalanced Support Vector Machines

“…Experiments indicate this method possess of the ability to detect known viruses and distinguish unknown viruses. In addition, it extracts code automatically and the virus condition code it generates has low error rate, which improves the rate of coverage of antigen preferably and decreases false alarm rate [15] . As indicated above, it is a practicable method for computer virus detection.…”