基于GA-PSO混合算法的最小属性约简  

Minimal attribute reduction algorithm based on GA-PSO

作　　者：吕振中[1] 薛惠锋[1] 钟明[1] 刘欢[1] 

机构地区：[1]西北工业大学自动化学院,西安710072

出　　处：《计算机工程与应用》2012年第1期53-56,共4页Computer Engineering and Applications

基　　金：西北工业大学科技创新基金(No.2008KJ02042)

摘　　要：属性约简是粗糙集理论中的核心问题,为有效进行属性的最小约简,将一种GA-PSO混合算法应用于属性约简。该算法在保证寻优能力的前提下,增加群体的多样性,避免陷入局部最优,同时,在适应度函数中加入罚函数。实验结果证明该算法能有效地进行属性约简,取得良好的约简结果。Attribute reduction is a key point of rough set theory. In order to get minimal subsets of attributes, this paper uses a GA-PSO mixed algorithm applying to attribute reduction. This algorithm in the premise of ensuring optimal ability, increases the diversity of population, and avoids being trapped in the local optimum, in the meantime, adds the penalty function in the fitness function. The experiment results show that it not only keeps the ability in getting reduction but also deduces the number of attribution, and it can obtain the prime effect.

关 键 词：粗糙集 属性约简 粒子群算法和遗传算法融合的混合算法(GA-PSO) 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]