came from. Similarly, another panel member requested that he be allowed to review the information that we had acquired about mission planning phases. A tracking system like HyperKAT enables developers to answer questions and respond to requests that affect system validity (and perhaps continuing project funding).
In a lesser sense HyperKAT can be used during a knowledge acquisition session. For example, after one of the initial knowledge acquisition sessions it may be prudent for the knowledge engineer and domain expert to work together and define, illustrate, and discuss the key terms and concepts that are pulled from the session review into the developing knowledge document. Such sessions not only save the knowledge engineer time and research but also can spark clarification of knowledge or the offering of new knowledge that may be stimulated by the reflection that this interactive task requires.
Furthermore, HyperKAT enables the ongoing development of a knowledge document that can be used to consolidate basic domain information. In turn, the knowledge document can be used as a training tool. For example, as knowledge engineers leave the project and are replaced by new personnel, the knowledge document provides a wealth of information that has proven to be important to the project. Reviewing knowledge document material can help new knowledge engineers come up to speed in the domain faster than if they were left to research, request information from multiple sources, and digest information on their own. Customers or clients may also recognize the value of a knowledge document. It represents a hard copy of vast amounts of domain expertise that they might want to share as a reference document for new employees.
Acknowledgments. The author is indebted to Christopher Westphal of the Institute for Defense Analyses, who as a member of the HyperKAT design team was responsible for programming HyperKAT and providing the screens for the figures in this chapter. Special thanks go to Robert Hoffman, who has worn his editor's hat so well during the chapter development process. The FRESH and Pilot's Associate program referred to in this chapter are DARPA-sponsored contracts and part of the Strategic Computing Program. The views expressed in this chapter about those programs are those of the author and do not reflect the views or opinions of DARPA, Texas Instruments, the Navy, or the Air Force.
Akscyn R., Yoder E., & McCracken D. ( 1988). The data model is the heart of interface design. Proceedings of CHI '88 (pp. 138-244). New York: Association for Computing Machinery.
Boose J. ( 1985). A knowledge acquisition program for expert systems based on personal construct psychology. International Journal of ManMachine Studies, 23, 495-525.
Boose J., & Bradshaw J. ( 1987). Expertise transfer and Complex Problems: Using AQUINAS as a knowledge acquisition workbench for expert systems. International Journal of Man-Machine Studies, 26, 495-525.