摘要
借鉴生物免疫学的有关理论,该文阐述了用于人工免疫系统的免疫优势概念,包括抗体免疫优势和抗原免疫优势.进一步地,构造了不同的免疫优势算子,并结合抗体克隆选择机理,提出了一种新的混合人工免疫系统算法一免疫优势克隆算法.0-1背包问题和复杂函数优化问题的仿真试验表明,新算法具有处理类似复杂问题的能力,且性能要优于相应的遗传算法.
Inspired by the biology immune system, the concept of Immunodominance in immunology is introduced to Artificial Immune System (AIS) in this paper. In AIS, the concept includes two parts, antibody-immunodominance and antigen- immunodominance. Based on the clonal selection theory, a new hybrid AIS algorithm, namely IrnmunoDominance Clone Algorithm (IDCA), with antibody-immunodominance is put forward. Compared with gene algorithms, IDCA is shown to be an evolutionary strategy capable of solving complex machine learning tasks, like 0-1 knapsack problem.
出处
《电子与信息学报》
EI
CSCD
北大核心
2004年第12期1918-1924,共7页
Journal of Electronics & Information Technology
基金
国家自然科学基金(60133010和60372045)资助项目
关键词
克隆
免疫优势
人工免疫系统
遗传算法
<Keyword>Clone, Immunodominance, Artificial immune system, Genetic algorithms
