摘要
针对人工免疫系统在处理多维向量分类时存在的初始抗体数量“爆炸”问题,对人工免疫系统的反向选择算法进行改进,研究了初始抗体的生成机理,并在初始抗体的生成条件中加入新的约束,降低了人工免疫系统所需的初始抗体数量,解决了多维向量的分类问题,在实际应用中有着很强的推广应用价值。
Artificial immune system is a new computational intelligence approach which uses ideas gleaned from immunology and has advantages in performing a wide range of tasks in various areas of research. However, the immune negative algorithm would cause a large number of the antibodies solving multi-di- mension data classification. Then in this paper, the immune negative algorithm was improved and will be applied in the muhi-dimension data classification. The results prove that the improved algorithm can solve the problem using a small number of the antibodies.
出处
《军械工程学院学报》
2009年第3期63-67,共5页
Journal of Ordnance Engineering College
基金
项目来源：国家自然科学基金资助（50705097）
关键词
人工免疫系统
反向选择算法
算法改进
初始检测器
artificial immune system
negative algorithm
algorithm improvement
initial antibodies
