Enhancing AINE (Artificial Immune Network) and Vertebrate Immune System in Cyber Security

Abstract

The main goal of the paper is to examine and to improve the anomaly detection function of artificial 
immune systems, specifically the negative selection algorithm and other self/non-self recognition techniques. This 
research investigates different representation schemes for the negative selection and proposes new detector 
generation algorithms suitable for such representations. Accordingly, different representations are explored: hyper
rectangles (which can be interpreted as rules), fuzzy rules, and hyper-spheres. Four different detector generation 
algorithms are proposed: Negative Selection with Detection Rules (NSDR, an evolutionary algorithm to generate 
hypercube detectors), Negative Selection with Fuzzy Detection Rules (NSFDR, an evolutionary algorithm to 
generate fuzzy-rule detectors), Real-valued Negative Selection (RNS, a heuristic algorithm to generate hyper
spherical detectors), and Randomized Real-valued Negative Selection (RRNS, an algorithm for generating hyper
spherical detectors based on Monte Carlo methods). Also, a hybrid immune learning algorithm, which combines 
RNS (or RRNS) and classification algorithms is developed. This algorithm allows the application of a supervised 
learning technique even when samples from only one class (normal) are available. Different experiments are 
performed with synthetic and real world data from different sources. The experimental results show that the 
proposed representations along with the proposed algorithms provide some advantages over the binary negative 
selection algorithm. The most relevant advantages include improved scalability, more expressiveness that allows the 
extraction of high-level domain knowledge, non-crisp distinction between normal and abnormal, and better 
performance in anomaly detection.