An Immune-inspired Approach to Learning and Classification

F. Azuaje (UK)


GA, classification, immune system.


The immune system exhibits interesting properties from an information processing point of view. This study applies basic immunology principles to perform classification tasks. A genetic algorithm provides the basis for the development of sample discrimination and aggregation processes. These mechanisms are combined to construct classifiers for symbolic data. Different aspects related to the learning and classification processes are evaluated. These methods may represent useful alternatives to address complex classification applications, such as concept learning in the absence of counter examples.

Important Links:

Go Back