Knowledge acquisition and refinement for a domain-specific expert system
1990 (English)Licentiate thesis, comprehensive summary (Other academic)
Tools for developing expert systems (knowledge-based systems) exist and are being used increasingly. However, many such tools have poor support for knowledge base management. Three aspects are essential in the development of knowledge base management tools, namely knowledge representation, knowledge acquisition and knowledge refinement.
This thesis discusses knowledge base management tools and reports an implementation for knowledge acquisition and refinement in a decision support system for artificial ventilation. The tool (KA VE) is based on a domain model and has facilities for entering, editing and refinement of the domain knowledge base. Morever an attached simulator and a knowledge editor support verification and refinement. KA VE also supports inconsistency and redundancy checking in the knowledge base.
Prelirninary experiences show that the tool makes knowledge verbalization faster than is possible without computer support. The graphical interface and the rule description format have a good readability and have been used by domain experts to validate and tune the artificial ventilation knowledge base.
Place, publisher, year, edition, pages
Linköping: Linköpings universitet , 1990. , 22 + 4 papers p.
Linköping Studies in Science and Technology. Thesis, ISSN 0280-7971 ; 229
Medical and Health Sciences
IdentifiersURN: urn:nbn:se:liu:diva-29513Local ID: LiU-Tek-Lic 1990:29ISBN: 91-7870-671-8OAI: oai:DiVA.org:liu-29513DiVA: diva2:250328
1990-06-01, Föelasmngssalen, IMT, plan 11, Lab Il-byggnaden, Campus US, Linköpings universitet, Linköping, 13:15 (English)