Fuzzy regression using asymmetric fuzzy coefficients and fuzzified neural networks

✍ Scribed by Hisao Ishibuchi; Manabu Nii

• Publisher: Elsevier Science
• Year: 2001
• Tongue: English
• Weight: 277 KB
• Volume: 119
• Category: Article
• ISSN: 0165-0114
• DOI: 10.1016/s0165-0114(98)00370-4

✦ Synopsis

In this paper, ÿrst we explain several versions of fuzzy regression methods based on linear fuzzy models with symmetric triangular fuzzy coe cients. Next we point out some limitations of such fuzzy regression methods. Then we extend the symmetric triangular fuzzy coe cients to asymmetric triangular and trapezoidal fuzzy numbers. We show that the limitations of the fuzzy regression methods with the symmetric triangular fuzzy coe cients are remedied by such extension. Several formulations of linear programming problems are proposed for determining asymmetric fuzzy coe cients from numerical data. Finally, we show how fuzziÿed neural networks can be utilized as nonlinear fuzzy models in fuzzy regression. In the fuzziÿed neural networks, asymmetric fuzzy numbers are used as connection weights. The fuzzy connection weights of the fuzziÿed neural networks correspond to the fuzzy coe cients of the linear fuzzy models. Nonlinear fuzzy regression based on the fuzziÿed neural networks is illustrated by computer simulations where Type I and Type II membership functions are determined from numerical data.