Artificial Immune System using Multi-Level Negative Selection Approach to Anomaly Detection

Natural immune system (NIS) provides a rich source of inspiration for computer security in the age of the Internet. The Artificial Immune System (AIS) is one of the promising techniques that seek to capture some aspects of the natural immune system. One of the major algorithms to implement the AIS is the Negative Selection (NS) algorithm. The paper proposes a new algorithm called MultiLevel Negative Selection (MLNS) which is an improved variant of the original algorithm. The proposed algorithm is compared with the original negative selection algorithm. Data from the international DARPA data set used to train and test the feasibility of the two algorithms. According to the experimental results, the proposed algorithm outperformed the original negative selection algorithm with higher detection rate while keeping a comparable false alarm rate. Furthermore, the number of detectors needed in case of the proposed algorithm is far less than the number of detectors if a single scale detector is used as in the case of the original negative selection algorithm.