Decision fusion of redundant knowledge based on rough set

The important research contents of rough set are classification and reduction, the purpose of which is to get rule knowledge and realize accuracy decision. The metrics for rule set is presented and the properties of a rule set are showed in general, which play an important role for decision of redundant knowledge. Based on the basic theory of model integration rough set, a formal model for multi-repositories is provided. One repository is regarded as a decision model. By building a model integration procedure decision fusion is realized.