Incompleteness in knowledge bases

  • Authors:
  • Hector J. Levesque

  • Affiliations:
  • Dept. of Computer Science, University of Toronto, Toronto, Ontario

  • Venue:
  • Proceedings of the 1980 workshop on Data abstraction, databases and conceptual modeling
  • Year:
  • 1980

Quantified Score

Hi-index 0.00

Visualization

Abstract

The first topic for discussion at the High Level Abstraction Workshop was What should be modelled? that is, what aspects of the world (slice of reality / enterprise) must be dealt with in a high level conceptual model. I would like to address this question from the point of view of incomplete knowledge bases. Although my remarks pertain mainly to AI applications, I suspect that any application using an information system of some sort will eventually have to face similar issues. The knowledge bases required in many applications can be characterized by a lack of complete information about the world of interest.