Artificial immune system

Artificial Immune Systems (AIS) are computer algorithms inspired by the principles and processes of the mammalian immune system.

At present artificial immune algorithms can be broadly categorised into two subgroups: those using the clonal selection theory and those using the immune network theory as their main inspiration. The algorithms typically exploit the immune system's charateristics of learning and memory to solve a problem.

History

The first works in the field of AIS came in 1986 when Farmer et al proposed a model for the immune network theory [1] and Hoffmann, motivated by a problem regarding neural networks, explored the similarities and differences between the nervous and immune systems [2]. Research into AIS began in earnest in the mid 90s in the realm of computer security for computer virus and network intrusion detection [3].

Uses

Artificial Immune Systems have been used to solve problems in a huge variety of domains. Some examples include:

E-mail classification (link)
Hardware fault tolerance (link)
Software testing (link)
Memory for context aware systems (link)

[1] Farmer, J.D., N.H. Packard, and A.S.Perelson, The immune System, Adaptation and machine learning. Phisica, 1986. 22D: p. 187-204.
[2] Hoffman, G.W., A Neural Network Model Based on the Analogy with the Immune System. Journal of Theoretical Biology, 1986. 122: p. 33-67.
[3] S. Forrest, A.S. Perelson, L. Allen, R. and Cherukuri. Self-Nonself Discrimination in a Computer. In Proceedings of the 1994 IEEE Symposium on Research in Security and Privacy, Los Alamitos, CA: IEEE Computer Society Press (1994)