不完备邻域决策粗糙集的最小化代价属性约简算法

目前的决策粗糙集研究主要集中在完备离散型信息系统,很少有对不完备连续型数据进行研究,考虑这一问题,提出一种不完备邻域决策粗糙集模型。首先在不完备连续型数据中引入了不完备邻域关系,然后利用该二元关系对传统的决策粗糙集进行重构,一种称之为不完备邻域决策粗糙集的模型被提出,同时基于决策代价原则,进一步地提出了最小化决策代价的属性约简算法。最后通过实验表明了所提出的算法具有更高的属性约简性能。

Minimum cost attribute reduction algorithms for incomplete neighborhood decision-theoretic rough set

The current research on decision-theoretic rough sets mainly focuses on the complete discrete information system,and few studies on incomplete continuous data.Considering this problem,this paper proposed an incomplete neighborhood decision-theoretic rough set model.Firstly,it introduced incomplete neighborhood relation into incomplete continuous data,and then reconstructed the traditional decision-theoretic rough sets by the binary relation,proposed a model called incomplete neighborhood decision-theoretic rough sets.At the same time,based on the principle of decision cost,it proposed an attribute reduction algorithm to minimize the decision cost.Finally,experiments show that the proposed algorithm has better performance in attribute reduction.