Abstract: On the basis of rough sets, by considering both the dynamic characteristics of the set [X] and the statistics information in the knowledge database, this paper constructs bidirectional transfer probabilistic PS-rough sets model, discusses properties and theorems of PS-rough sets, proves that PS-rough sets is the further extension of S-rough sets and Z. Pawlak rough sets, S-rough sets and Z. Pawlak rough sets are the special cases of PS-rough sets. Comparing with S-rough sets, the approximation quality of the dynamic set [X*] in PS-rough sets is relatively increased and then the decision accuracy is improved. The efficiency of PS-rough sets is shown by an example.

Key words: S-rough sets, bidirectional transfer probabilistic PS-rough sets, probabilistic approximation space

摘要： 在粗糙集基础上，既考虑集合[X]的动态特性，又考虑知识库中的统计信息，构建了概率近似空间上的双向迁移PS-粗糙集模型，讨论了PS-粗糙集的性质及相关定理，证明了PS-粗糙集是S-粗糙集和Z.Pawlak粗糙集的进一步扩展，S-粗糙集和Z.Pawlak粗糙集是PS-粗糙集的特例。与S-粗糙集相比，PS-粗糙集的动态集合[X*]的近似精度得到相对提高，从而提高了决策精度。通过实例验证了PS-粗糙集的有效性。

关键词: S-粗糙集, 双向迁移概率PS-粗糙集, 概率近似空间