Implementation of Fuzzy Logic Method and Certainty Factor for Diagnosis Expert System of Chronic Kidney Disease

Abstract

Problem analysis from the development of technology that was originally conducted manually, it can now conduct systematically using computerization. An expert system can solve one problem analysis in diagnosing a disease like chronic kidney disease. Fuzzy logic and certainty factor are expert system methods that are often used. The data used in this research was Chronic Kidney Disease, which was obtained from the UCI dataset. The system was developed using the Laravel PHP framework programming language and MySQL database. The system's development used the waterfall method, which was analyzing user needs to the system, conducting design of the system, coding, and testing the system if it achieves what was expected. The combination of fuzzy logic and certainty factor methods worked with several stages, namely fuzzification (CFuser), rule base formation for CF, calculating CFexpert, Calculating CF values, the combination of CF values, Finding CFmaximum. The accuracy level of the system generated from 400 data was obtained 92.25% accuracy for the fuzzy logic method, 97.25% accuracy for the certainty factor, 99% accuracy method for the combination of fuzzy logic and certainty factor methods. While the kappa value for the fuzzy logic method, certainty factor, and the combination of the two methods were respectively 0.84, 0.94, 0.98.