In this paper, we explore the application of Opt-AiNet, an immune network approach for search and\noptimisation problems, to learning qualitative models in the form of qualitative differential equations.\nThe Opt-AiNet algorithm is adapted to qualitative model learning problems, resulting in the proposed\nsystem QML-AiNet. The potential of QML-AiNet to address the scalability and multimodal search space\nissues of qualitative model learning has been investigated. More importantly, to further improve the\nefficiency of QML-AiNet, we also modify the mutation operator according to the features of discrete\nqualitative model space. Experimental results show that the performance of QML-AiNet is comparable\nto QML-CLONALG, a QML system using the clonal selection algorithm (CLONALG). More importantly,\nQML-AiNet with the modified mutation operator can significantly improve the scalability of QML and is\nmuch more efficient than QML-CLONALG.
Loading....