Ontologies and the Knowledge Acquisition Bottleneck
Mihai Boicu, Gheorghe D. Tecuci, Bogdan Adrian Stanescu, Catalin Balan, Elena Popovici · International Joint Conference on Artificial Intelligence · 2001
Ontologies and information sharing have a major role to play in the development of knowledge-based agents and the overcome of the knowledge acquisition bottleneck. This paper supports this claim by presenting an approach to ontology specification, import, and development that is part of Disciple-RKF. Disciple-RKF is a theory, methodology, and learning agent shell for the rapid development of knowledge-based agents by subject matter experts, with limited assistance from knowledge engineers. The Disciple approach has been subject of intensive evaluations, as part of DARPA’s “High Performance Knowledge Bases” and “Rapid Knowledge Formation” programs, demonstrating very good results.