A Discrimination Based Artificial Immune System for Classification

This paper presents a new artificial immune system for classification. It is named a discrimination based artificial immune system (DAIS). It is based on the principle of self-nonself discrimination by T cells in the human immune system. Ability of a natural immune system to distinguish between self and nonself molecules is applicable for classification in a way that one class is distinguished from other. We demonstrate the behavior of DAIS and show this system is efficient for artificial datasets and also for real world datasets. It has comparable performance to other classifier systems, while it needs much less memory