This article is cited in 6 papers

Simulation of Behavior and Intelligence

Optimal piecewise-linear approximation algorithms for complex dependencies

E. V. Bauman, A. A. Dorofeyuk, G. V. Kornilov

Institute of Control Sciences, Russian Academy of Sciences

Abstract: Methods for fuzzy piecewise-linear approximation based on the general variational approach to data classification analysis are described. The underlying idea is to partition the space of variables (input parameters) into domains in each of which a function (dependence) that is complicated in the whole space is approximated by a linear function.

Presented by the member of Editorial Board: F. T. Aleskerov

Received: 25.12.2003

 English version: Automation and Remote Control, 2004, 65:10, 1667–1674

Bibliographic databases:

• 
• 
• 
•