An explanation facility for today's expert systems
- 1 January 1989
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Expert
- Vol. 4 (1) , 26-36
- https://doi.org/10.1109/64.21897
Abstract
The authors discuss explanation facility types found in most expert systems and shells used today, referring to these as practice systems, since they represent current practice in expert system work. Practice systems include nonresearch expert systems (being developed every day in industry) as well as nonresearch expert system shells. They maintain that an extensive explanation facility meeting the obligations of a practice system's role can be implemented using only practice technology. They discuss the relationship of journalism to explanation, showing that explanation facility tasks (roles) on current practice expert systems and shells resemble a newswriter's tasks, as both involve presenting users or readers with accurate, objective, noninteractive accounts of events. In news reporting, events are pieces of news. In expert-system explanation, events are actions taken by expert systems. The authors present a journalistic explanation facility, Joe, which is part of AGNESS (a generalized network-based expert system shell), under development at the University of Minnesota.Keywords
This publication has 10 references indexed in Scilit:
- Radon transform orientation estimation for rotation invariant texture analysisIEEE Transactions on Pattern Analysis and Machine Intelligence, 2005
- Enhanced Maintenance and Explanation of Expert Systems Through Explicit Models of Their DevelopmentIEEE Transactions on Software Engineering, 1985
- An expert system for a resource allocation problemCommunications of the ACM, 1985
- The role of frame-based representation in reasoningCommunications of the ACM, 1985
- Strategic explanations for a diagnostic consultation systemInternational Journal of Man-Machine Studies, 1984
- Deep versus compiled knowledge approaches to diagnostic problem-solvingInternational Journal of Man-Machine Studies, 1983
- XPLAIN: a system for creating and explaining expert consulting programsArtificial Intelligence, 1983
- The epistemology of a rule-based expert system —a framework for explanationArtificial Intelligence, 1983
- A Functional Description of ANALYZE: A Computer-Assisted Analysis System for Linear Programming ModelsACM Transactions on Mathematical Software, 1983
- Tutoring rules for guiding a case method dialogueInternational Journal of Man-Machine Studies, 1979