Case-Based Reasoning as a Paradigm for Heuristic Search

Abstract

Case-based reasoning (CBR) systems utilize known solutions to specific problems (a case base) in order to guide problem solving behavior in the face of new problems. The CBR approach to problem solving can easily be exploited to achieve learning capabilities if we incorporate new knowledge into the case base as we acquire additional problem-solving experience. In this paper we will show how the case-based reasoning approach to problem solving can be used to heuristically limit exhaustive tree searches. As an important aspect of this approach, we will also show how the concept of coarse indices can be used to expand the power of specific problem/solution pairs to order to saturate a search space. We have applied these ideas to an implementation of the 8-puzzle in an effort to illustrate how case-based reasoning fosters a class of weak methods for general problem solving and machine learning.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Oct 01, 1987
Accession Number
ADA249334

Entities

People

  • Wendy G. Lehnert

Organizations

  • University of Massachusetts Amherst

Tags

Communities of Interest

  • Autonomy
  • Biomedical
  • Energy and Power Technologies

DTIC Thesaurus Topics

  • Algorithms
  • Artificial Intelligence
  • Computer Programming
  • Computer Programs
  • Computer Science
  • Computers
  • Databases
  • Engineering
  • Expert Systems
  • Information Science
  • Intelligent Systems
  • Lisp Programming Language
  • Machine Learning
  • Normal Distribution
  • Random Walk
  • Test And Evaluation
  • Universities

Fields of Study

  • Computer science

Readers

  • Artificial Intelligence

Technology Areas

  • AI & ML
  • AI & ML - Machine Learning Algorithms
  • Space