A fast computational algorithm of a fuzzy reasoning using the product-sum-gravity method

✍ Scribed by Makoto Ohki; Syuhei Hayashi; Masaaki Ohkita

• Publisher: John Wiley and Sons
• Year: 2000
• Tongue: English
• Weight: 258 KB
• Volume: 31
• Category: Article
• ISSN: 0882-1666
• DOI: 10.1002/(sici)1520-684x(200003)31:3<40::aid-scj5>3.0.co;2-y

✦ Synopsis

This paper proposes a fast computational algorithm for the reasoning called the algebraic product-sum-gravity method, among the computational methods for fuzzy reasoning. In the proposed method, the reasoning expression by the product-sum-gravity method is modified and the integral operation is extracted. By calculating the integral expression analytically beforehand, the operation in the fuzzy reasoning execution can be replaced by a simple algebraic operation. In addition, by imposing a certain constraint in defining the antecedent membership function, the number of rules to be considered in the computation can be reduced. By recomposing the computational procedure, the expression providing the reasoning output can be represented by the fast computational flow. By calculating using that fast computational flow, the fuzzy reasoning by the product-sum-gravity method is executed at high speed.