CoalSORT: a knowledge-based interface to an information retrieval system. Final report
The aim of the CoalSORT project is to determine the most appropriate ways to apply Al techniques to facilitate and extend the use of bibliographic databases in coal technology. The project's main task is to design and implement a prototype which provides an intelligent interface to a bibliographic database. The intelligence of the prototype is embodied in a frame-based semantic network which is a representation of an expert's domain knowledge, especially in its cognitive organization. Achieving optimal use of this intelligence depends on how effective the system is at communicating partial descriptors which express a match between a user's information need and a document's information content. The system communicates with the user through menus, windows for displaying and keeping track of information, and the selection of key phrases as search terms. A number of features were added to enable the interface to present the shape or organization of the categorial network system. This organization plays an important role in delineating the meaning of key concepts communicated to the user. The adequacy of this knowledge representation will be tested by appropriate users. 23 refs., 38 figs.
- Research Organization:
- Carnegie-Mellon Univ., Pittsburgh, PA (USA). Dept. of Computer Science
- DOE Contract Number:
- AC22-83PC62260
- OSTI ID:
- 5380974
- Report Number(s):
- DOE/PC/62260-T1; ON: DE86012714
- Country of Publication:
- United States
- Language:
- English
Similar Records
Some problems and proposals for knowledge representation
Fact retrieval for the 1980s