COMPOSITIONAL RULE OF CHAIN INFERENCE IN COMPUTATIONAL INTELLIGENCE PROBLEMS

Abstract

The incorporation of imprecise, linguistic information into logical deduction processes is a significant issue in computational intelligence. Throughout the literature, we can find all sorts of intelligent inference schemes acting under imprecision; common to most approaches is their reliance on if-then rules of the kind “IF X is A THEN Y is B”, where A and B are fuzzy sets (FS) in given universes U and V. While the FS-based theory of approximate reasoning is surely a well-established and commonly applied one, there is still for further expanding the expressiveness of the formalism. One such improvement can be obtained by using picture fuzzy sets (PFS), in which the sets A and B are picture fuzzy sets in the corresponding universes U and V. In this paper, we will contribute to the further development of the picture fuzzy logic (PFL) by presenting some new classes of implication operators in PFL and firstly defining the Compositional Rule of Chain Inference (CRCI) in a PFL setting. The new chain inference procedures should be applied in computational intelligence problems