This specification method of the consequent class of fuzzy rules has been used in a few studies since early 1990s [ 26].
The consequent class C q is specified by identifying the class with the maximum confidence: (3) where M is the number of classes.
Its consequent class C q and rule weight CF qare specified from compatible training patterns with A q in the following heuristic manner.
It should be noted that the same consequent class as in Equations 2 and 3 is obtained when we use the support of the fuzzy rule A q ⇒ Class h instead of the confidence degree.
In the above rule, x = (x1,…, x n) is the n-dimensional input vector, A qi is an antecedent fuzzy set for the i – th input variable, C q is a consequent class, and CF q is a certainty degree (i.e., rule weight).
It is shown that the creep of metal matrix composites is controlled by creep of the matrix alloys and there is a consequent division of the creep behavior of the composites into two types, class M and class A, as in the creep of solid solution alloys.
For the disease prediction application, the rules of interest are called fuzzy class association rules (FCARs), meaning that they have only one consequent: the class.
