A Human Immunity Inspired Algorithm to Detect Infections in a Computer Program

K. Wawryn, P. Widulinski (Koszalin Univ. of Techn., Poland)

In this paper, an algorithm to detect computer program modifications possibly made by malware has been proposed. The algorithm has two steps: receptor generation using a method based on templates, and anomaly detection which utilizes a negative selection algorithm inspired by the human immune system. The proposed algorithm has been implemented and tested. Experimental results are presented, analyzed and concluded.