This paper presents a detailed parametric study of an immunology inspired algorithm applied to a classification problem. The algorithm, known
as the Artificial Immune Network Algorithm (AINet), has its effectiveness
examined through various parameter settings on an Iris flower dataset. We
analyze the performance of the parameters given to AINet and their impact on
its classification accuracy as well as considering acceptable execution times.
The results demonstrate that certain parameters can significantly optimize the
algorithm’s accuracy without overly compromising the computational efficiency
compared to others. This study provides valuable data for the application of
artificial immunology algorithms in classification problems.
