A Study on Analysis of Malicious Codes Similarity Using N-Gram and Vector Space Model

Abstract

The development of Information and Communication gives a lot of convenience in our lives, but on the other hand, the new cyber threat like viruses, computer intrusions and so on also increases. Especially, malicious codes such as viruses over the internet are distributed in large quantities. Several studies using reverse engineering or virtual machine have been carried out to analyze malicious codes, but studies on the analysis of the implications or the similarities between malicious codes have not been carried out in variety. Therefore, we suggest the analysis of similarity using N-Gram and Vector Space Model in this paper. The results shows that we can distinguish malicious codes from others using this method.