Fuzzy Sets and Systems, ( ISI ), Volume (481), No (1), Year (2024-4) , Pages (108890-108905)

Title : ( Robust hybrid learning approach for adaptive neuro-fuzzy inference systems )

Authors: Ali Nik Khorasani , Ali Mehrizi , Hadi Sadoghi Yazdi ,

Citation: BibTeX | EndNote


The Adaptive Neuro-Fuzzy Inference System (ANFIS) is a regression model that uses fuzzy logic and neural networks, making it suitable for modeling the uncertainty of regression problems. However, the non-robust loss function in ANFIS\\\\\\\'s hybrid learning algorithm can make it susceptible to the direct effects of noise and outliers. This article introduces a new procedure that uses robust loss functions to enhance the hybrid learning performance against noise and outliers. Specifically, a new robust loss function that can completely ignore outliers is devised, and a set of robust loss functions with mathematical relations is suggested. The proposed approach is evaluated on real-world problems, including weather forecasting and stock market prediction, and the results show that it can reduce the Mean Square Error (MSE) in regression. Moreover, the new procedure enables the use of different loss functions based on the application.


, Adaptive Neuro-Fuzzy Inference System, Loss function, Hybrid learning, Robustness, Weather Forecasting, Stock Market Forecasting
برای دانلود از شناسه و رمز عبور پرتال پویا استفاده کنید.

author = {Nik Khorasani, Ali and Mehrizi, Ali and Sadoghi Yazdi, Hadi},
title = {Robust hybrid learning approach for adaptive neuro-fuzzy inference systems},
journal = {Fuzzy Sets and Systems},
year = {2024},
volume = {481},
number = {1},
month = {April},
issn = {0165-0114},
pages = {108890--108905},
numpages = {15},
keywords = {Adaptive Neuro-Fuzzy Inference System; Loss function; Hybrid learning; Robustness; Weather Forecasting; Stock Market Forecasting},


%0 Journal Article
%T Robust hybrid learning approach for adaptive neuro-fuzzy inference systems
%A Nik Khorasani, Ali
%A Mehrizi, Ali
%A Sadoghi Yazdi, Hadi
%J Fuzzy Sets and Systems
%@ 0165-0114
%D 2024