In this paper, M-estimators, where M stands for maximum likelihood, used in robust regression theory for linear parametric regression problems will be generalized to nonparametric maximum likelihood fuzzy neural networks (MFNNs) for nonlinear regression problems. Emphasis is put particularly on the robustness against outliers. This provides alternative learning machines when faced with general ...

fuzzy linear programming problem occur in many elds such as mathematical modeling, control theory and management sciences, etc. in this paper we focus on a kind of linear programming with fuzzy numbers and variables namely fully fuzzy linear programming (fflp) problem, in which the constraints are in inequality forms. then a new method is proposed to ne the fuzzy solution for solving (fflp). ...

The well known fuzzy measures, λ-measure and P-measure, have only one formulaic solution. A multivalent fuzzy measure with infinitely many solutions based on Pmeasure was proposed by our previous work, called completed L-measure. In this paper, a further improved fuzzy measure, called extensional completed L-measure, is proposed. This new fuzzy measure is proved that it is not only an extension...

We propose a discriminative fuzzy clustering maximum a posterior linear regression (DFCMAPLR) model adaptation approach to compensate the acoustic mismatch due to speaker variability. The DFCMAPLR approach adopts the MAP criterion and a discriminative objective function to estimate shared affine transform and fuzzy weight sets, respectively. Then, through a linear combination of the calculated ...

The aim of this paper is to give a. sketch of a foundation of statistical analysis with fuzzy data, extending the methodology of Mathematical Stat,istics. For t.his purpose it will be discussed to equip the fuzzy sample spaces with suitable a-algebras and metrics. As the main application the parameter estimation in linear regression models with crisp parameters und fuzzy-valued variables will b...

In recent years, the fuzzy linear regression (FLR) approach is widely applied in the quality function deployment (QFD) to identify the vague and inexact functional relationships between the customer requirements and the engineering characteristics on account of its advantages of objectiveness and reality. However, the h value, which is a vital parameter in the proceeding of the FLR model, is us...

The accuracy of regression model may have some warp when the observation data set exists outlier. We study to lessen the outlier effect in predicting GDP growth on the basis of fuzzy linear regression model in this paper. The result is eff ectively verified the performance of the model.

in this paper, we use parametric form of fuzzy number and we converta fuzzy linear system to two linear system in crisp case. conditions for the existence of a minimal solution to $mtimes n$ fuzzy linear equation systems are derived and a numerical procedure for calculating the minimal solution is designed. numerical examples are presented to illustrate the proposed method.