A Knowledge Acquisition Tool for the Intelligent Processing of Materials.

Abstract

Despite recent advances that have been made in expert system building technologies the process of obtaining knowledge from human experts and coding it into a computing system remains fraught with difficulties. The Knowledge Acquisition Tool (KAT) currently being developed at the BDM Corporation is designed to address this knowledge acquisition bottleneck in the domain of materials processing. KAT offers the knowledge engineer a variety of modes in which to query a materials expert on different aspects of a given materials domain. In the Clarification Mode, the expert specifies the entire process in terms of an AND/OR graph. This structure is converted to a flowchart in the Prediction Mode, where the expert is asked to predict problems that can occur at different points in the process, and to specify limiting conditions. Finally, the Diagnosis Mode provides a dynamic step-through of the process model as defined by the expert, so that s/he can verify the specification of the process and redesign as necessary. Progress made thus far on the KAT project includes a) completion of the overall system design; b) implementation of an initial version of the Clarification Mode; c) partial completion of the Prediction Mode; and d)detailed consideration of KAT's generality to domains other than materials processing.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Dec 31, 1987
Accession Number
ADA188982

Entities

People

  • Teresa A. Blaxton

Organizations

  • Braddock Dunn & McDonald

Tags

Communities of Interest

  • Ground and Sea Platforms
  • Sensors

DTIC Thesaurus Topics

  • Acquisition
  • Artificial Intelligence
  • Composite Materials
  • Computer Programming
  • Computer Programs
  • Computers
  • Corporations
  • Engineering
  • Engineers
  • Expert Systems
  • Materials
  • Materials Engineering
  • Materials Processing
  • Operating Systems
  • Simulations
  • Software Development

Readers

  • Artificial Intelligence
  • Computer Science.
  • Theoretical Analysis.