摘要：This paper studies fuzzy rule-based systems (FRBS) design using genetic algorithms (GA). GA parameters such as population size, crossover probability, and mutation rate can play a vital role in enhancing the performance of FRBS. The experimental results and the comparison of different parameters show the improvement in the performance of FRBS. An FRBS is entirely defined by fuzzy membership function (MF), and it is crucial to select the precise MFs. The manual process of obtaining an accurate boundary of MF is a pretty complex task. This research designed a system that will help obtain an optimal MF boundary using GA. The developed models are compared with the FRBS without GA. The result obtained is based on the models' simulation on two different datasets, i.e. iris and appendicitis. The presented results show the improvement in the performance of the GA-based FRBS as compared to FRBS without GA for classification problems.
关键词：Fuzzy Rule-Based System;Genetic Algorithm;Tuning of Membership Function;Crossover;Mutation
