Fuzzy inferencing using single-antecedent fuzzy rules

The output of a fuzzy cognitive map (FCM) is the summation of the products of its individual input variables and their relative weights to each node. This additive nature is the main hindrance to the implementation of fuzzy theory in FCM, as conventional fuzzy rules rely on the mapping of the input spaces to the output space via the intersection configuration of fuzzy rules. On the other hand, fuzzy rules are well known for the combinatorial rule explosion problem. We present a methodology that allows the use of single-antecedent fuzzy rules instead of multiple-antecedent fuzzy rules. This methodology thus can be implemented in FCM and at the same time eliminate altogether the problem associated with the intersection of fuzzy rules. Our proposed method ensures transparency of the rules, unlike in conventional data-driven fuzzy systems where they can become ambiguous