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

A.A.A. Youssif, A.Z. Ghalwash, and S.A. Mohamed (Egypt)


Artificial Immune System, Intrusion Detection, Negative Selection, and 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 an immunological algorithm based on the Negative Selection Algorithm and the Clonal Selection technique, called the Multi-Level Negative Selection (MLNS). The proposed algorithm is compared with the previous work of the AIS. Data from the international DARPA data set is used to train and test the feasibility of the new algorithm. The recorded experimental results show that the proposed algorithm outperforms the previous work and a higher detection rate is achieved (96%:94.5%). Meanwhile, a comparable false alarm rate is attained (1%:0.9%). A remarkable advantage, of the proposed algorithm, is the noticeable reduction in the number of detectors needed to achieve the stated results since it comes down to nearly a quarter (22.3%) of those generated with the previously used single scale detector.

Important Links:

Go Back