Title: A fuzzy functional network for nonlinear regression problems

Authors: Matteo Gaeta; Vincenzo Loia; Stefania Tomasiello

Addresses: Department of Computer Engineering, Electrical Engineering and Applied Mathematics, University of Salerno, 84084, Fisciano, Italy ' Department of Computer Science, University of Salerno, 84084, Fisciano, Italy ' Department of Computer Engineering, Electrical Engineering and Applied Mathematics, University of Salerno, 84084, Fisciano, Italy

Abstract: This paper introduces functional networks in a fuzzy environment. We formally define a fuzzy functional network for application in some regression problems and prove a sufficient condition on the solution of our regression model. Several numerical examples, based on simulated data and literature data, show the good performance of our approach: solutions seem not sensitive to outliers, without using large training datasets.

Keywords: functional networks; fuzzy sets; nonlinearity; fuzzy logic; nonlinear regression.

DOI: 10.1504/IJKESDP.2014.069290

International Journal of Knowledge Engineering and Soft Data Paradigms, 2014 Vol.4 No.4, pp.351 - 370

Published online: 09 May 2015 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article