针对现实生活中的大部分信息是缺省的和属性取值不确定性的情况,首先引入了区间灰集概念,并在其基础上对不完备信息系统进行了研究,讨论了区间灰集信息系统基于优势关系下的属性约简,提出了基于区间灰集的可能度与优势度,进而利用其对对象进行排序,并通过引入阀值,使排序达到了人的主观要求与客观实际更高层次上的一致,最后通过实例说明如何求属性约简集以及利用约简集排序是合理的.
Considering the reality that most of the information systems are the default values and the situation that many attributes are uncertainty in the real world ,a new rough set model named interval-set-value was proposed in this paper. On the basis of it,the incomplete information system was studied. The attribute reduction under dominance relations was discussed and the possibility degree and inclusion degrees was proposed based on the interval-set-value model. The properties of them were obtained respectively. Then the objects were sorted on the basis of these models. Finally,the examples were given to show how to obtain attribute reduction sets, and it is reasonable to use it for ranking. Further more,a higher consistency for both subjective and objective requirement to sort objects was got by introducing a threshold value.