An Analysis of Evolutionary Methodology for Interpretable Logical Fuzzy Rule-Based Systems

Abstract

Machine modeling approach entails constructing dynamical prototypes explaining the performance of real networks from measurable data using analytical models and technologies. Using Fuzzy Logic (FL) necessitates a trade-off between interpretability and efficiency. According to essential theories and system identification techniques, achieving precise and also human-comprehensible FL plays is fundamental and plays a crucial role. Prior to the introduction of soft computing, however, FL model makers' primary priority was reliability, bringing the resultant FL nearer to black-box frameworks like neural networks. Fortunately, the Infinite-valued modelling scientific world has returned to its roots by exploring design strategies that address the interpretability and accuracy tradeoff. Because of their intrinsic versatility and capacity to examine several optimization criteria simultaneously, the application of evolutionary FL control has been greatly expanded. This paper is a study of the most typical evolutionary Infinite-valued technologies that use Mamdani Infinite-valued rule-based approaches to produce interpretable logical Fuzzy Rule-Based Systems (FRBSs), which are highly interpretable.