From Multiple Classification RDR to Configuration RDR

Access & Terms of Use
metadata only access
Abstract
Ripple Down Rules (RDR) is a knowledge acquisition method for knowledge based systems (KBS) which facilitates incremental acquisition of knowledge and ensures that the previous performance of the KBS is not degraded by the incremental addition of the new knowledge. This approach is now well established for single classification tasks and more recently has been extended to multiple classification tasks. This paper describes the further extension of the approach to configuration tasks. The test domain for this study is the configuration of ion chromatography methods in analytical chemistry.
Persistent link to this record
DOI
Link to Publisher Version
Additional Link
Author(s)
Ramadan, Ziad
Byrnes-Preston, Philip
Le-Gia, Thong
Chellen, Vija
Compton, Paul
Mulholland, Mary
Hibbert, D. Brynn
Haddad, Paul
Kang, Byeong
Supervisor(s)
Creator(s)
Editor(s)
Translator(s)
Curator(s)
Designer(s)
Arranger(s)
Composer(s)
Recordist(s)
Conference Proceedings Editor(s)
Other Contributor(s)
Corporate/Industry Contributor(s)
Publication Year
1998
Resource Type
Conference Paper
Degree Type
UNSW Faculty