Abstract: The current research status of rough set reduction is reviewed, and a new rough set reduction algorithm based on the quantum genetic algorithm(QGA) is proposed, which is free from the disadvantages in genetic algorithms from the respect of recursion and convergence performance. With a new type of discernibility matrix adopted, the algorithm exhibits excellent capabilities in attribute reduction of inconsistent decision table; a new parametric configuration scheme for the fitness function is proposed, which can be used as a reduct admissibility criterion. It has been validated experimentally that the proposed algorithm is superior to the conventional GA-based algorithm in terms of convergence and speed.

Key words: rough set, quantum genetic algorithm, attribute reduction

摘要： 分析了粗糙集属性约简的研究现状，针对遗传算法求取属性约简中存在的迭代次数多、收敛较慢的问题，提出了基于量子遗传算法的粗糙集属性约简的新方法。该方法中利用一种新的区分矩阵与量子遗传算法结合，能够实现相容/不相容决策表的属性约简；同时，文中提出了一种适应度函数的参数设定的新方法，使之能够直接对约简进行有效判定。实验数据表明：该算法在收敛性和速度等方面优于基于遗传算法的属性约简算法。

关键词: 粗糙集, 量子遗传算法, 属性约简

CLC Number: 

• TP312