A structure learning method for concise fuzzy systems

This paper presents a structure learning method for fuzzy systems following our previous work on a Structure Evolving Learning Method for Fuzzy Systems (SELM) and an Evolving Construction Scheme for Fuzzy Systems (ECSFS). Here we extend our previous work to a structure learning method for fuzzy systems which results in more concise systems. We analyse and compare the proposed concise structure learning strategies in terms of three aspects: (1) how to detect the splitting points for the structure learning process; (2) how to set a starting point for the fuzzy system; (3) how the proposed method is applied to Mamdani and TS fuzzy systems.

1. Introduction

Since the invention of fuzzy sets and fuzzy systems by Zadeh four decades ago, fuzzy systems have been successfully applied in many different areas. These successes are grounded on the advantages of fuzzy systems [1]–[7]. Firstly, fuzzy systems can be identified by combining human knowledge and information from data. Secondly, the transparency and interpretability of fuzzy rules and systems make them easy to understand and use in applications. Lastly, fuzzy systems are universal approximators, that is, they can approximate any continuous nonlinear function to any degree of accuracy and therefore can be used in the solution of general complex and nonlinear problems.