Dealing with Uncertainty in Fuzzy Inductive Reasoning Methodology

The aim of this research is to develop a strat­ egy of reasoning under uncertainty in the context of the Fuzzy Inductive Reasoning methodology. This methodology allows the prediction of systems behavior by means of two different schemes. The first one corre­ sponds to a pattern prediction scheme, based exclusively on pattern rules. The second one corresponds to a purely Sugeno infer­ ence system, i.e. Sugeno prediction scheme. The Sugeno fuzzy rules are automatically ex­ tracted from the pattern rules producing a compact representation of the system mod­ elled. In this paper a mixed pattern/fuzzy rules scheme is studied to deal with uncer­ tainty in such a way that the best of both perspectives is used. The proposed scheme is applied to a real biomedical system, i.e. the central nervous system control of the cardio­ vascular system.