A novel rule antecedent structure and its identification for fuzzy models

The existing combinatorial antecedent structure in fuzzy models makes them suffer from ldquothe curse of dimensionalityrdquo. In this work, a novel rule antecedent structure is proposed to design an efficient fuzzy model by using fewer rules. The new rule antecedent only uses nonlinear variables. Additionally, the proposed rule antecedents are expressed as ellipsoids covering the underlying local regions, which make spatial coverage more efficient.