Abstract: High accuracy of the description to injection mould repair schemes is required and excess inducted rules are obtained when the knowledge is inducted by Pawlak rough set, therefore, the feasibility of the conventional rough set in practical application is poor. By the combination of feature and concept hierarchy model and the definition of innovation about concept, the hierarchical representation of repair schemes and knowledge is put forward. The repair schemes can be reconstructed hierarchically by the calculation of innovation about concept which represents the fuzzy description of them and then feature decision tables are achieved. By the abstraction of innovation about concept from the feature decision tables, feature fuzzy similar matrix is constructed and the feature decision table is divided into some fuzzy equivalent subsets by introducing confidence level vector. The algorithm of feature reduction and knowledge induction based on fuzzy rough set is put forward by defining single or multiple fuzzy feature equivalence relations. A case study indicates that the feasibility of this method and the universality of the inducted knowledge are both prompted obviously.

Key words: Fuzzy rough set, Innovation, Knowledge induction