Fuzzy classification model based on decision-theoretic rough set

A new fuzzy classification rnodel is proposed The proposed rnodel uses a decision-theoretic rough set to design the structure of a fuzzy classification rnodel The fuzzy C-rneans clustering algorithrn is used to transforrn the continuous attributes into discretized ones and to partition the fuzzy input space A heuristic attribute reduction algorithrn based on a two-step search strategy deals with the discretized decision table to rernove redundant-condition attributes Then, concise decision rules are extracted The rules of the fuzzy classification rnodel are obtained according to the extracted decision rules The fuzzy classification rules of the proposed rnodel have clear physical rneaning and a sirnplified structure Moreover, a learning algorithrn is no longer needed to optirnize the pararneters of the fuzzy rnodel Finally, the proposed rnodel is cornpared with sorne existing classification algorithrns by experirnents using sorne UCI data sets The experirnent results show that the proposed rnodel is effectie.