论文部分内容阅读
属性约简是不协调覆盖决策系统中的核心问题之一。该文在不协调覆盖决策系统中提出了具有非负性和单调性的限制信息熵、限制条件信息熵与限制互信息,在此基础上讨论了协调集、约简以及核的等价判定定理;然后,在不协调覆盖决策系统中提出了一种完备的具有多项式时间复杂度的启发式属性约简算法,并通过评估房屋出售情况的例子说明该算法的可行性与有效性。
Attribute reduction is one of the core issues in the uncoordinated coverage decision system. This paper proposes a non-negative and monotonic restricted information entropy, a restricted information entropy and a restricted mutual information in the uncoordinated coverage decision-making system. On the basis of this, we discuss the coordination sets, reductions and the equivalent judgments of the kernel Then, a complete heuristic attribute reduction algorithm with polynomial time complexity is proposed in the uncoordinated coverage decision-making system. The feasibility and validity of the algorithm are illustrated by evaluating the sale of houses.