What's in a deep model?: a characterization of knowledge depth in intelligent safety systems

  • Authors:

  • David Klein;Tim Finin

  • Affiliations:

  • Computer and Information Science, University of Pennsylvania, Philadelphia, PA;Computer and Information Science, University of Pennsylvania, Philadelphia, PA

  • Venue:
  • IJCAI'87 Proceedings of the 10th international joint conference on Artificial intelligence - Volume 1
  • Year:
  • 1987

Quantified Score

Hi-index 0.00

Visualization

Abstract

While one can characterize deep and shallow models at a high level of abstraction and contrast their relative merits in a general way, this provides little direction for knowledge engineering. In particular, the field lacks a clear definition of 'knowledge depth' and lacks guidelines regarding the appropriate depth of models for a given application, in this paper we provide a very simple operational definition of knowledge depth' and use it to examine the opportunities for varying depth in Intelligent safety systems. The paper illustrates a domain-independent mode of analysis for examining progressively deeper models of expertise, and sketches some domain-specific guidelines for constructing intelligent safety systems. We draw upon examples from the domains of nuclear reactor management, chemical plant control, and management of computer installation operations.