Fuzzy Parameters Estimation via Hybrid Methods

Abstract

Fuzzy regression analysis is one of the most widely used statistical techniques which represents the relation between variables. In this paper, the crisp inputs and the symmetrical triangular fuzzy output are considered. Two hybrid algorithms are considered to fit the fuzzy regression model. In this study, algorithms are based on adaptive neuro-fuzzy inference system. The results are derived based on the V-fold cross validation, so that the validity and quality of the suggested methods can be guaranteed. Finally, using the numerical examples, the performance of the suggested methods are compared with the other ones, such as linear programming (LP) and quadratic programming (QP). Based on examples, hybrid methods are verified for the prediction.

Optimal design of adaptive neuro-fuzzy inference system using PSO and ant colony optimization for estimation of uncertain observed values

“…In addition, linear programming was used for the consequence parameters prediction. Moreover, these algorithms were compared with the method proposed by Danesh (2018) which was based on adaptive fuzzy inference system and linear programming (FWLP). The study showed that the proposed methods had fewer errors than the LP and FWLP methods, and were further verified by the predictions using simulation and practical examples.…”

Regression parameters prediction in data set with outliers using neural network

“…In a study of Takagi and Sugeno [33], the method was presented for identifying a system using its input-output data. Also in 2009 and 2014, Dalkilic and Apaydin [7,8] used the ANFIS model to analyze switching regression and estimate the fuzzy regression parameters, and in 2016, Danesh et al [9,10] used the ANFIS model to predict fuzzy regression model. Generally for real-world applications, data sets often contain multiple variables as well as noise or outliers that are inconsistent with the other data.…”