DECO3RUM: A Differential Evolution learning approach for generating compact Mamdani fuzzy rule-based models

作者：

Highlights：

• A novel evolutionary Fuzzy Rule-based System for modeling problems is proposed.

• The system utilizes Differential Evolution as its learning algorithm.

• DECORUM was validated using 24 real-world datasets.

• Additionally, one dataset from the domain of soil spectroscopy was examined.

• Experimental results indicate that DECORUM achieves notable results.

摘要

•A novel evolutionary Fuzzy Rule-based System for modeling problems is proposed.•The system utilizes Differential Evolution as its learning algorithm.•DECORUM was validated using 24 real-world datasets.•Additionally, one dataset from the domain of soil spectroscopy was examined.•Experimental results indicate that DECORUM achieves notable results.

论文关键词：Fuzzy Rule-based System (FRBS),Evolutionary fuzzy systems,Differential Evolution,Soil spectroscopy,Fuzzy Token Competition,Genetic Cooperative - Competitive Learning (GCCL)

论文评审过程：Received 20 December 2016, Revised 13 April 2017, Accepted 14 April 2017, Available online 15 April 2017, Version of Record 10 May 2017.

论文官网地址：https://doi.org/10.1016/j.eswa.2017.04.026