关键词: 粗糙集, 粗关系数据库, 查询模型

Abstract: In this paper，the storage，index and search of uncertain data in a rough relational database（RRDB） are studied by combining database theory and rough sets.First，two storing techniques are proposed.One is to store equivalence classes with regard to the values of attributes by using an adjacency list.The other is to store records of RRDB by using an orthogonal list.Secondly，based on Hamming distance and the given clustering method，one way to index uncertain data in RRDB is put forward.Thirdly，an approach to calculating the similarity between the data entered by users and the data being queried in RRDB based on the concepts of upper and lower approximation of rough sets is presented.Fourthly，a querying model of RRDB is constructed and an algorithm for querying uncertain data in RRDB is presented.Finally，a real world example is illustrated and results shows that the proposed algorithm is useful and effective.

Key words: rough sets, rough relational database, query model