gIBIS: a hypertext tool for exploratory policy discussion
ACM Transactions on Information Systems (TOIS)
Model-based reasoning: troubleshooting
Exploring artificial intelligence
From ideas and arguments to hyperdocuments: travelling through activity spaces
HYPERTEXT '89 Proceedings of the second annual ACM conference on Hypertext
Exploring representation problems using hypertext
HYPERTEXT '87 Proceedings of the ACM conference on Hypertext
Explanations in Knowledge Systems: Design for Explainable Expert Systems
IEEE Expert: Intelligent Systems and Their Applications
Automated Knowledge Acquisition for Strategic Knowledge
Machine Learning
Machine understanding of devices: causal explanation of diagnostic conclusions
Machine understanding of devices: causal explanation of diagnostic conclusions
The elements of computer credibility
Proceedings of the SIGCHI conference on Human Factors in Computing Systems
Explaining collaborative filtering recommendations
CSCW '00 Proceedings of the 2000 ACM conference on Computer supported cooperative work
Establishing and maintaining long-term human-computer relationships
ACM Transactions on Computer-Human Interaction (TOCHI)
Hi-index | 0.01 |
That intelligent systems need an explanatory capability if they are to aid or support human users has long been understood. A system which can justify its decisions generally obtains improved user trust, greater accuracy in use and offers embedded training potential. Extensive work has been done to provide rule-based systems with explanatory interfaces, but little has been done to provide the same benefits for model-based systems. We develop an approach to organizing the presentation of large amounts of model-based data in an interactive format patterned after a model of human-human explanatory and argumentative discourse. Portions of this interface were implemented for Honeywell's model-based Flight Control Maintenance and Diagnostic System (FCMDS). We conclude that sufficient information exists in a model-based system to provide a wide range of explanation types, and that, the discourse approach is a convenient, powerful and broadly applicable method of organizing and controlling information exchange involving this data.