The Reformulation Model of Expertise.

Abstract

This research develops a methodology for implementing a class of expert problem-solving programs which must create models of the problems they are given before they can apply their expert knowledge to those problems -- i.e., given a problem in their domain of expertise, these programs 'reformulate' the problem description in terms of their built-in abstract models of the domain. The implementation methodology provides the following capabilities: (1) Pieces of expert knowledge can be added to the program in a fairly independent manner without regard to their control structure implications; (2) Programs can delve through irrelevant information in the problem description to pick out only the facts needed for problem-solving; (3) Programs can model the input problem at various levels of detail, as required by the problem-solving task; and (4) Programs can tailor their expert models to the problem at hand. This is accomplished by controlling the general procedure which maps pieces of the expert model into pieces of the input with feedback information taken from the model of the input so far.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Dec 01, 1976
Accession Number
ADA035397

Entities

People

  • William Scott Mark

Organizations

  • Massachusetts Institute of Technology

Tags

Communities of Interest

  • Biomedical
  • Cyber
  • Engineered Resilient Systems
  • Human Systems
  • Weapons Technologies

DTIC Thesaurus Topics

  • Artificial Intelligence
  • Computational Science
  • Computer Programming
  • Computer Programs
  • Computer Science
  • Computers
  • Databases
  • Electrical Engineering
  • Employment
  • Expert Systems
  • Information Systems
  • Language
  • Materials
  • Mathematical Models
  • Psychology
  • Quality Control
  • Reasoning

Readers

  • Adaptive Control and Estimation with Uncertainty in Dynamic Systems.
  • Artificial Intelligence
  • Educational Psychology