A Function Model for Fuzzy Knowledge Representation and Inference

The solutions of many real-world problems require reasoning under fuzzy logic.Fuzzy reasoning is well-known techniques in control system,expert systems and various fields of artificial intelligence.In this paper,we discussed the representation of fuzzy knowledge on the basis of fuzzy sets and fuzzy inference theory,and used the member-ship functions to discript the fuzzy logic variables.We then show that the inference mechanism on conditional possibility distributions leads to valid results.The role of the predicate modefiers in fuzzy reasoning was also discussed.