Redundancy criteria for linguistic fuzzy rules

The elimination of redundant rules in a given system is considered to be an important step towards its simplicity, clarity, and comprehensibility. In this paper, we return to the problem of redundancies in systems of linguistic fuzzy IF-THEN rules based on the perception-based logical deduction inference method. We generalize the existing criteria for detecting redundant rules in these systems to general ones, and elaborate a new algorithm for an automatic search and deletion of redundant linguistic fuzzy rules. The criteria are provided in a theoretically formulated form with full proofs. Furthermore, we present the final algorithm and demonstrative examples that show the positive impact of the new algorithm.