摘要
离散化在粗糙集的实际应用中有重要的意义,它使得粗糙集有更广泛的应用范围。粗糙集中的离散化要求在保持原来决策系统的不可分辩关系情况下,用尽量少的断点进行离散化。文章针对该问题提出了一种遗传算法,将最小断点集作为优化目标,并构造一个新的算子来保证所选断点能保持原决策系统的不可分辩关系,同时相应地采用自适应变异策略。最后对实例进行了求解。
Discretization of attributes with real values is an important problem the applying of rough set,and it widens the scope of using rough set.Discretization based on rough set is to seek possibly minimum number of discrete inter-vals,and at the same time it should not weaken the indiscernibility ability of the original decision system.A genetic al-gorithm for this problem is proposed.The minimum set of cuts is the goal of optimizing.A new operator is constructed to maintain the discernibility of the cuts selected.The self-adapted mutation strategy is constructed accordingly.And the re-sult of the algorithm on a real data set is showed.
出处
《计算机工程与应用》
CSCD
北大核心
2003年第8期13-14,28,共3页
Computer Engineering and Applications
基金
国家自然科学基金资助(编号:69703011)
关键词
粗糙集
离散化
断点
Rough set,Discretization,Cuts
