View article

Multi–granulation is an improvement of the classical rough set theory since it uses a family of binary relations instead of of a single indiscernibility relation for the constructing of approximation. In this paper, the multi–granulation rough set approach is further generalized into fuzzy environment. A family of fuzzy T–similarity relations are used to define the optimistic and pessimistic fuzzy rough sets respectively. The basic properties about these fuzzy rough sets are then discussed. Finally, the relationships among single relation based fuzzy rough set, optimistic and pessimistic multi–granulation fuzzy rough sets are addressed.