Improving Artificial Immune System Performance: Inductive Bias and Alternative Mutations

摘要

Research works related to the Artificial Immune System (AIS) and their applications have been extensively reported during the last decade. In this work, we proposed an inductive bias heuristic called neighbourhood improvement within the classical AIS for improving its performance. We also demonstrated alternative mutation mechanisms for cloning the elite antibodies. Computational experiments using the proposed heuristic and mechanisms to find the near optimal solutions of travelling salesman problems were conducted. The results obtained from the modified AIS were compared with those obtained from other metaheuristics. It was found that the performance of the modified AIS adopting the proposed heuristic and mechanisms outperformed the conventional AIS and other metaheuristics.