Genetically optimized linguistic models
Keun-Chang Kwak · 2008
This paper is concerned with linguistic models (LM) optimized by genetic algorithm (GA) and developed their comprehensive framework. The main characteristics of LM are user-centric and inherently dwell upon collections of highly interpretable and user-oriented entities such as information granules. The objective of this paper is to present an organization of overall optimization process and come up with a specification of several evaluation mechanisms of the performance of the models. The underlying design tool guiding the development of LM revolves around the augmented version of fuzzy clustering known a context-based fuzzy c-means. The optimization design based on GA determines the number of cluster generated by each linguistic context and fuzzification factor related to information granules in the input and output process. The experimental study comes with coagulant dosing process in a water purification plant. Furthermore we contrast the performance of genetically opimized LM in comparison with other radial basis function networks and LM itself.