Algoritmos para Identificación de Modelos Borrosos Interpretables (Data-Driven Identification Algorithms for Automatic Determination of Interpretable Fuzzy Models)

Juan Contreras Montes (epcontrerasj@ieee.org)1, Roger Misa Llorca (rmisa@electrica.cujae.edu.cu)2, Lisbeth Urueta Vivanco (lisbethu@curn.edu.co)3


1Escuela Naval Almirante Padilla
2Instituto Superior Politécnico José Antonio Echeverría
3Coproración Universitaria Rafael Núñez

This paper appears in: Revista IEEE América Latina

Publication Date: Sept. 2007
Volume: 5,   Issue: 5 
ISSN: 1548-0992


Abstract:
This article presents a new methodology to obtain fuzzy models linguistically interpretable from input and output data. The proposed methodology includes the class determination and rules generation algorithms, as long as the partition sum-1 of the input variables: shape, number and distribution of the fuzzy sets. The most promising issue on our proposal is represented by the equilibrium between precision and interpretability of the model. Applications to well-known problems and data sets are presented and compared with the results of other authors using different techniques.

Index Terms:
identification, clustering, fuzzy model, interpretability   


Documents that cite this document
This function is not implemented yet.


[PDF Full-Text (359)]