Knowledge extraction using the FARB

Eyal Kolman*, Michael Margaliot

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingChapterpeer-review

Abstract

The FARB-ANN equivalence immediately yields a representation of the IO mapping of a given trained ANN in a symbolic form. The fuzzy rules and their parameters are determined by the network structure and parameters. This transforms the knowledge embedded in the network into a set of symbolic rules stated using natural language. If necessary, the FARB can be further simplified to increase its comprehensibility.

Original languageEnglish
Title of host publicationKnowledge-Based Neurocomputing
Subtitle of host publicationA Fuzzy Logic Approach
EditorsEyal Kolman, Michael Margaliot
PublisherSpringer Berlin Heidelberg
Pages41-57
Number of pages17
ISBN (Print)9783540880769
DOIs
StatePublished - 2009

Publication series

NameStudies in Fuzziness and Soft Computing
Volume234
ISSN (Print)1434-9922

Fingerprint

Dive into the research topics of 'Knowledge extraction using the FARB'. Together they form a unique fingerprint.

Cite this