关键词: 粗糙集, 正区域, 递归计算

Abstract: The computing of positive regions is very important in the application of rough sets. Traditional computing methods are more complex. So based on an indepth study of the rough set theory, a equivalent method for computing positive regions is proposed. The algorithm classifies the samples in universe step by step and determines whether those samples belong to the positive region and deletes the samples from the universe to reduce computational complexity and gets the unpositive region set. Then, a method of recursively computing the positive region is proposed by analyzing the influence of the increased attributes to the set of known positive regions. So the positive region of an increased attribute system can be gotten by analyzing the unpositive region. And the theoretical analysis and experimental results show that this new method is much more efficient and useful in computing.

Key words: rough set, positive region, recursive computing