ABSTRACT
This chapter introduces a problem statement and general methodology of design and tuning of fuzzy rules in the rough and fine stages. It presents the basic models necessary for the fuzzy logic inference and rough tuning of fuzzy rules by the method of pairwise comparison. The chapter also presents fine tuning as a problem of optimization, mathematical models for fuzzy inference quality evaluation, and some results of computer experiments for the fuzzy models with continuous and discrete consequents, respectively. It provides to the application of the proposed approach in building a fuzzy expert system for the differential diagnosis of ischemia heart disease. The chapter describes how to build the fuzzy rule-based system for the differential diagnosis of IHD by the proposed methodology. From the formal point of view, the problem of fuzzy model creation for medical diagnosis can be considered as a problem of nonlinear object identification with multiple inputs and a single output.
