有约束指导的粗集决策表属性约简算法

Constraint guidance reduction method of rough sets decision-making table

粗集理论通过对原始决策表的约简获取规则知识,其核心部分是属性约简.本文在结合两种基本属性约简算法优点的基础上提出了有约束指导的属性约简算法,并给出了算法的详细步骤.该算法通过专家经验的约束指导避免了对属性之间随机组合情况的搜索,可以提高求解速度.仿真试验验证了该方法的有效性和优越性.应用该算法处理一组生产调度数据以获取调度规则,结果显示能较好的与生产实际相吻合,这进一步验证了算法的实用性.

Rough set theory acquires rules knowledge through the reduction of the original decision table, and its core part is reduction of attributes. This paper presents a reduction algorithm of attributes with constraint guidance, which is based upon the combination of two basic merits of reduction algorithm of attributes, and proposes the detailed steps of the algorithm. The algorithm avoids the search for random composition among attributes via constraint guidance of rules of thumb, and increases computing speed. Simulation tests verify the effectiveness and superiority of the algorithm. As an example, the algorithm was used to acquire a set of scheduling rules from production scheduling data. The results demonstrate that theses rules can better coincide with the actual production. This is a further test of the practicality of the algorithm.