Incorporation of Adapted Real World Tournament Selection into Artificial Immune Recognition System

摘要

The resource competition phase of the Artificial Immune Recognition System (AIRS) incorporates a selection mechanism with a high selective pressure and loss of diversity. This selection mechanism generates premature memory cells and decreases the classification accuracy. In this study, the Real World Tournament Selection (RWTS) method is incorporated in resource competition phase of AIRS to tackle this limitation. Some experiments are conducted to evaluate the accuracy of new algorithm, named RWTSAIRS. Algorithms were tested on benchmark datasets of UCI machine learning repository and RWTSAIRS achieved better classification accuracy in all cases.