eFSLab: Developing Evolving Fuzzy Systems from Data in a Friendly Environment



A software lab is presented to support the development of fuzzy systems from data (data-driven approach) avoiding redundancy and unnecessary complexity in the obtained membership functions, in order to give some semantic meaning to the results. On-line mechanisms for merging membership functions and rule base simplification are implemented improving interpretability and transparency of the produced fuzzy models, allowing the minimization of redundancy and complexity of the models during their development, contributing to the transparency of the obtained rules. The application, developed in Matlab environment, and public under GNU license, is applied to one benchmark problem- the Box-Jenkins time series prediction- with illustrative results.


Soft Computing


10th European Control Conference, August 2009

Cited by

Year 2011 : 1 citations

 Adaptive kernel smoothing regression using vector quantization
FM Pouzols, A Lendasse - Evolving and Adaptive Intelligent …, 2011 -

Year 2010 : 2 citations

 Evolving fuzzy optimally pruned extreme learning machine: A comparative analysis
FM Pouzols, A Lendasse - Fuzzy Systems (FUZZ), 2010 IEEE …, 2010 -

 Evolving fuzzy optimally pruned extreme learning machine for regression problems
FM Pouzols, A Lendasse - Evolving Systems, 2010 - Springer