What are fuzzy rules and how to use them

Didier Dubois (1, 2) , Henri Prade (1, 2)

1 IRIT-ADRIA - Argumentation, Décision, Raisonnement, Incertitude et Apprentissage
2 CNRS - Centre National de la Recherche Scientifique

Fuzzy rules have been advocated as a key tool for expressing pieces of knowledge in “fuzzy logic”. However, there does not exist a unique kind of fuzzy rules, nor is there only one type of “fuzzy logic”. This diversity has caused many a misunderstanding in the literature of fuzzy control. The paper is a survey of different possible semantics for a fuzzy rule and shows how they can be captured in the framework of fuzzy set and possibility theory. It is pointed out that the interpretation of fuzzy rules dictates the way the fuzzy rules should be combined. The various kinds of fuzzy rules considered in the paper (gradual rules, certainty rules, possibility rules, and others) have different inference behaviors and correspond to various intended uses and applications. The representation of fuzzy unless-rules is briefly investigated on the basis of their intended meaning. The problem of defining and checking the coherence of a block of parallel fuzzy rules is also briefly addressed. This issue has been neglected in the fuzzy control literature although it looks important for validation purposes.

Keywords: Approximate reasoning; Fuzzy rules; Knowledge representation; Gradualness; Certainty; Possibility; Exceptions; Consistency

Domains: Artificial Intelligence [cs.AI]

Didier Dubois, Henri Prade. What are fuzzy rules and how to use them. Fuzzy Sets and Systems, 1996, 84, dedicated to the Memory of Professor Arnold Kaufmann (2), pp.169--185. ⟨10.1016/0165-0114(96)00066-8⟩. ⟨hal-04039815⟩