Reconstructive expert system explanation

  • Authors:
  • Michael R. Wick;William B. Thompson

  • Affiliations:
  • Univ. of Wisconsin-Eau Claire, Eau Claire;Univ. of Utah, Salt Lake City

  • Venue:
  • Artificial Intelligence
  • Year:
  • 1992

Quantified Score

Hi-index 0.00

Visualization

Abstract

Existing explanation facilities are typically far more appropriatefor knowledge engineers engaged in system maintenance than for end-usersof the system. This is because the explanation is little more than atrace of the detailed problem-solving steps. An alternative approachrecognizes that an effective explanation often needs to substantiallyreorganize the actual line of reasoning and bring to bear additionalinformation to support the result. Explanation itself becomes a complexproblem-solving process that depends not only on the actual line ofreasoning, but also on additional knowledge of the domain. This paperpresents a new computational model of explanation and argues that itresults in significant improvements over traditional approaches.—Authors' Abstract