A Combination of Negative Selection Algorithm and Artificial Immune Network for Virus Detection

This paper proposes a combined approach of Negative Selection Algorithm and Artificial Immune Network for virus detection. The approach contains the following stages: the first stage is data extraction and clustering. In the second stage, the negative selection algorithm is deployed to create the first generation of detectors. In the third stage, aiNet is used to improve detectors’ coverage and enhance the ability of detecting unknown viruses. Finally, generated detectors are used to computing danger level of files and a classifier is used to train them and test performance of the system. The experimental results show that in the suitable conditions, the proposed approach can achieve reasonably high virus detection rate.