Modified Clonal Selection Algorithm Based Classifiers

摘要

The biological immune system is an adaptive, complex and robust system that helps the body defend from foreign pathogens. Clonal Selection algorithm (CLONALG) is one of the many algorithms that have been inspired by the adaptive biological immunity of human being and animals. CLONALG has been applied in data mining, pattern recognition and optimization problems. The present paper presents a modified CLONALG based classifier algorithms. CLONALG has many steps and one of these steps is initializing the antibodies pool. The present paper has proposed a new approach to initialize the antibodies pool for classifier design and provides some tests and experiments to show the effectiveness of CLONALG classifier performance with randomized and antigen initializations.