Competitive Exception Learning Using Fuzzy Frequency Distributions

15 Pages Posted: 3 Feb 2003

See all articles by W. M. van den Bergh

W. M. van den Bergh

Erasmus University Rotterdam (EUR) - Erasmus School of Economics (ESE); Erasmus Research Institute of Management (ERIM)

J. van den Berg

Delft University of Technology, Faculty of Technology, Policy and Management, Section of ICT

Date Written: February 2000 5,

Abstract

A competitive exception learning algorithm for finding a non-linear mapping is proposed which puts the emphasis on the discovery of the important exceptions rather than the main rules. To do so,we first cluster the output space using a competitive fuzzy clustering algorithm and derive a fuzzy frequency distribution describing the general, average system's output behavior. Next, we look for a fuzzy partitioning of the input space in such away that the corresponding fuzzy output frequency distributions `deviate at most' from the average one as found in the first step. In this way, the most important `exceptional regions' in the input-output relation are determined. Using the joint input-output fuzzy frequency distributions, the complete input-output function as extracted from the data, can be expressed mathematically. In addition, the exceptions encountered can be collected and described as a set of fuzzy if-then-else-rules. Besides presenting a theoretical description of the new exception learning algorithm, we report on the outcomes of certain practical simulations.

Keywords: exception learning, fuzzy pattern recognition, competitive learning

JEL Classification: M, M11, R4, C6

Suggested Citation

van den Bergh, W. M. and van den Berg, J., Competitive Exception Learning Using Fuzzy Frequency Distributions (February 2000 5,). Available at SSRN: https://ssrn.com/abstract=370808

W. M. Van den Bergh (Contact Author)

Erasmus University Rotterdam (EUR) - Erasmus School of Economics (ESE) ( email )

Burgemeester Oudlaan 50
3062 PA Rotterdam, NL 3062 PA
Netherlands

Erasmus Research Institute of Management (ERIM)

P.O. Box 1738
3000 DR Rotterdam
Netherlands

J. Van den Berg

Delft University of Technology, Faculty of Technology, Policy and Management, Section of ICT ( email )

P.O. Box 5015
2600 GB Delft
Netherlands

HOME PAGE: http://tbm.tudelft.nl/index.php?id=30084&L=1

Do you have a job opening that you would like to promote on SSRN?

Paper statistics

Downloads
109
Abstract Views
1,300
rank
307,359
PlumX Metrics