A Detector Generating Algorithm for Intrusion Detection Inspired by Artificial Immune System

Artificial immune system (AIS) allows us to inspire several ideas for the design of computer intrusion detection. The standard of negative selection algorithm (NSA), offered by Stephanie Forrest in 1994, is one of the most common mechanisms in AIS that applied in anomaly detection for the similarity of its basic idea. One of the most operational improvements in the standard of NSA is how to generate effective detectors which play a significant role in self and nonself discrimination in intrusion detection system (IDS). In this paper, we offer an improvement to a detector generating algorithm to generate effective detectors which leads to improve the standard of NSA, which in turn leads to improve the NSA based anomaly intrusion detection. Experimental results show that the improved algorithm able to generate more effective detectors and keeping the space and time complexities better than in the standard of NSA. This leads to detecting the real-time intrusion with less false negative.