摘要
为了获得有效的属性最小相对约简,提出了一种基于自适应遗传算法的粗糙集知识约简算法。该算法将核引入遗传算法的初始群体来提高算法的性能,依照决策属性对条件属性的依赖度,在加强局部搜索能力的同时保持了该算法全局寻优的特性,并且对交叉概率和变异概率进行了新的设计。设计中既考虑到进化代数对算法的影响,又考虑到每代中不同个体适应度对算法的作用。最后通过两个经典算例进行了验证,无论在约简的准确性上,还是平均运行代数上都取得了较好的结果。
In order to get the reduction of attribute,the paper proposes a rough set attribute reduction algorithm based on AGA. The core is joined initial population in AGA in order to accelerate capability.According to the dependability of decision attribute to the condition attribute,it can but only obtain the capability of part searching,but also retain the peculiarity of all searching. The adaptive crossover probability and adaptive mutation probability are designed,considering the influence of every generation to algorithm and the effect of different individual fitness in every generation.Experimental results show that the accurate reduction and the average algebraic sum all obtain the preferable values.
出处
《计算机工程与应用》
CSCD
北大核心
2008年第15期1-3,11,共4页
Computer Engineering and Applications
基金
国家自然科学基金(the National Natural Science Foundation of China under Grant No.60474069)
关键词
粗糙集
知识约简
自适应遗传算法
交叉概率
变异概率
rough sets
knowledge reduction
Adaptive Genetic Algorithm(AGA)
crossover probability
mutation probability