A Novel Immunity-Based Anomaly Detection Method

Abstract

The human immune system consists of a complex set of cells and molecules that protect organs against infection. With many different kinds of lymphocytes (B cell, T cell and so on) distributing all over the human body, the human immune system can distinguish nonself from self and then eliminate nonself immediately. According to the principles of human immune system, a novel immunity-based anomaly detection method (NIAD) is presented. In NIAD, the formal definitions of self, nonself, detectors, immune tolerance, and etc., are given. Then, the quantitative description of the detector diversity is introduced to improve the generating efficiency of memory detectors, to reduce the number of memory detectors and to enlarge the coverage of nonself space. Furthermore, immune response is described. To determine the performance of NIAD, the experiments comparing with different anomaly detection methods, such as negative selection algorithm (NSM), multilevel immune learning algorithm (MILA), and variable sized detectors algorithm (V-detector), were performed. Experiments show that NIAD has a better performance than previous methods.