โ Scribed by Henk J. van Zuylen
No coin nor oath required. For personal study only.
Numerical programs are often difficult to use for people who were not involved in their development. Knowledge necessary for the use of the program can be included in a user interface. Often, further improvements are possible when the user interface gives the appropriate support. The user task can be modified by the introduction of a good user interface such, that less knowledge is necessary. The remaining knowledge can be inserted in a knowledge base, coupled to the user interface. The real improvement comes often from the fact that a good user interface makes it possible to work in a way which requires less knowledge.
The role of knowledge analysis and (cognitive) task analysis is indispensable. Prototyping starting from such an analysis gives a better prospect for few iteration and a good final system.
๐ SIMILAR VOLUMES
Although scientific computing traditionally involves numerical data, the need for symbolic representation is now recognized. In this context, we have developed a knowledge-based system coupled with a finite element library. It guides the user in the specification of the physical problem, finds a cor
In this article, we argue that a refinement tool capable of assisting the maintainer of a Knowledge-Based System (KBS) with updating and upgrading of its knowledge base can substantially expand the scope of his activities and prolong the life of the system. We show that refinement tasks involved in
Recent research in knowledge-based expert systems k)r VLSI tools" design has concentrated on floor-planning, placement, routing, and (:ell generation. The paper describes a knowledge-based scanning-line approach to VLSI layout compaction. This approach differs Jrom traditional algorithmic approaches