Design and implementation of Fuzzy Expert System using Fuzzy Assessment Methodology

This paper describes the design, implementation of fuzzy expert system for diagnosis of diabetes. The components of fuzzy expert system are fuzzification interface, Fuzzy assessment methodology and Defuzzification interface. Fuzzification interface converts the crisp values into fuzzy values. Fuzzy assessment methodology uses fuzzy operators, membership function, correlation fuzzy logic and probability to manage uncertainity in rules. Defuzzification interface converts the resulting fuzzy set into crisp values. To demonstrate the effectiveness of the proposed algorithm MATLAB Fuzzy Logic tool box is used for performance assessment. The result indicates that the fuzzy assessment methodology is very effective in improving the accuracy for diabetes application.