Empirical Analysis and Refinement of Expert System Knowledge Bases

Abstract

Scoring schemes for measuring expert system performance are reviewed. Rule-based classification systems and their error rates on sample data are considered. The authors present several models of measurement that are categorized by four characteristics-mutual exclusivity of classes, unique answers provided by the system, known correct conclusions for each case and use of confidence factors to weight the system's conclusions. An underlying model of performance measurement is critical in determining which scoring strategy is appropriate for a system and whether a comparison of different systems can be made.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Nov 30, 1988
Accession Number
ADA203607

Entities

People

  • Casimir A. Kulikowski
  • Sholom M. Weiss

Organizations

  • Rutgers University–New Brunswick

Tags

Communities of Interest

  • Human Systems

DTIC Thesaurus Topics

  • Acquisition
  • Artificial Intelligence
  • Automatic
  • Classification
  • Computer Science
  • Computers
  • Costs
  • Demonstrations
  • Diseases And Disorders
  • Errors
  • Expert Systems
  • Machine Learning
  • Measurement
  • Pattern Recognition
  • Recognition
  • Rule Based Systems
  • Supervised Machine Learning

Readers

  • Artificial Intelligence
  • Computational Modeling and Simulation
  • Instructional Design and Training Evaluation.