摘要
将最小属性约简问题转化为一个基于粒子群优化算法求解的多目标优化问题.引入基于表现型共享的适应度评价函数以提高多目标搜索算法的性能,对基本粒子群优化算法的位置更新公式进行修正使其能够有效应用于最小属性约简问题,并提出了一种用于求解该问题的二进制多目标粒子群优化算法.实验表明,本算法是有效的,并能一次运算获得多个最小属性约简.
In this paper,the minimum attribute reduction problem is transformed to a multi-objective optimization problem based on particle swarm optimization algorithm. A fitness function based on phenotype sharing is introduced to enhance the performance of the multi-objective search algorithm. The basic particle position updating formula is modified so that it can be effectively applied to the minimum attribute reduction problem. Finally,a binary multi-objective particle swarm optimization algorithm is proposed which allows us to find more minimum attribute reductions in a single run of the algorithm. Experiments show the effectiveness of the proposed algorithm.
出处
《福州大学学报（自然科学版）》
CAS
CSCD
北大核心
2010年第2期193-197,共5页
Journal of Fuzhou University(Natural Science Edition)
基金
教育部科学技术研究重点资助项目(206073)
关键词
最小属性约简
粒子群优化
算法
多目标优化
表现型共享
minimum attribute reduction
particle swarm optimization
algorithm
multi-objective optimization
phenotype sharing
