Multi-objective decision making: towards improvement of accuracy, interpretability and design autonomy in hierarchical genetic fuzzy systems

Presents fuzzy modeling as a multi-objective decision making problem considering accuracy, interpretability and autonomy as goals. The proposed approach assumes that these goals can be handled via corresponding single-objective /spl epsiv/-constrained decision making problems whose solution is produced by a hierarchical evolutionary process. The fitting, generalization, and interpretation characteristics of the resulting fuzzy models are discussed using a classification problem.

I. Introduction

Rule-based models are considered a linguistic representation of the mental model of a certain system by means of experience [1]. Fuzzy rule-based systems and models use fuzzy set theory to express expert knowledge in various domains [2]. In many applications, humans may be unable to extract knowledge out of massive amounts of numerical data, specially when the knowledge required is not easily available. This leads to the development of computer techniques to extract and represent knowledge in a fuzzy rule-based system. Data-driven fuzzy modeling, or fuzzy modeling (FM) for short, has attracted interest of many researches [3]–[7].