Solving Fuzzy Linear Regression with Hybrid Optimization

Authors:
M. H. Mashinchi;M. A. Orgun;M. Mashinchi
Affiliations:
Department of Computing, Macquarie University, Sydney, Australia 2109;Department of Computing, Macquarie University, Sydney, Australia 2109;Department of Statistics, Faculty of Mathematics and Computer Science, Shahid Bahonar University of Kerman, Iran
Venue:
ICONIP '09 Proceedings of the 16th International Conference on Neural Information Processing: Part II
Year:
2009

Citing 11
Cited 1

Evaluation of fuzzy linear regression models by comparing membership functions

Fuzzy Sets and Systems
Fuzzy regression methods—a comparative assessment

Fuzzy Sets and Systems
A Taxonomy of Hybrid Metaheuristics

Journal of Heuristics
A fuzzy linear regression model with better explanatory power

Fuzzy Sets and Systems - Information processing
A novel fuzzy linear regression model based on a non-equality possibility index and optimum uncertainty

Applied Soft Computing
An enhanced fuzzy linear regression model with more flexible spreads

Fuzzy Sets and Systems
An improvement on genetic-based learning method for fuzzy artificial neural networks

Applied Soft Computing
Regression with fuzzy random data

Computational Statistics & Data Analysis
Efficient hybrid methods for global continuous optimization based on simulated annealing

Computers and Operations Research
Linear and non-linear fuzzy regression: Evolutionary algorithm solutions

Fuzzy Sets and Systems
On cluster-wise fuzzy regression analysis

IEEE Transactions on Systems, Man, and Cybernetics, Part B: Cybernetics

Hybrid optimization with improved tabu search

Applied Soft Computing

Quantified Score

Hi-index	0.00

Visualization

Abstract

Fuzzy linear regression is an important tool to find the linear inexact relationship between uncertain data. We then propose a hybrid optimization method based on tabu search and harmony search as a potential way of solving fuzzy linear regression. The proposed method aims at finding a model without considering any mathematical constraints while reducing the error of the regression's model in comparison to other methods. The experimental comparison of the results for two classes of crisp input-fuzzy output and fuzzy input-fuzzy output data sets shows the superiority of the method over the existing ones.