Interactive classification: A technique for acquiring and maintaining knowledge bases

Journal Article · Wed Oct 01 04:00:00 EDT 1986 · Proc. IEEE; (United States)

DOI:https://doi.org/10.1109/PROC.1986.13642· OSTI ID:6905460

Finin, T W

The practical application of knowledge-based systems, such as in expert systems, often requires the maintenance of large amounts of declarative knowledge. As a knowledge base (KB) grows in size and complexity, it becomes more difficult to maintain and extend. Even someone who is familiar with the knowledge domain, how it is represented in the KB, and the actual contents of the current KB may have severe difficulties in updating it. Even if the difficulties can be tolerated, there is a very real danger that inconsistencies and errors may be introduced into the KB through the modification. This paper describes an approach to this problem based on a tool called an interactive classifier. An interactive classifier uses the contents of the existing KB and knowledge about its representation to help the maintainer describe new KB objects. The interactive classifier will identify the appropriate taxonomic location for the newly described object and add it to the KB. The new object is allowed to be a generalization of existing KB objects, enabling the system to learn more about existing objects.

• Research Organization: Dept. of Information Science, Univ. of Pennsylvania, Philadelphia, PA 19104-6389
• OSTI ID: 6905460
• Journal Information: Proc. IEEE; (United States), Journal Name: Proc. IEEE; (United States) Vol. 74:10; ISSN IEEPA
• Country of Publication: United States
• Language: English

Related Subjects

99 GENERAL AND MISCELLANEOUS
990210* -- Supercomputers-- (1987-1989)
ARTIFICIAL INTELLIGENCE
CLASSIFICATION
DATA COVARIANCES
ERRORS
EXPERT SYSTEMS
KNOWLEDGE BASE
MAINTENANCE