Planning Techniques for Rule Selection in Deductive Question-Answering,

Abstract

Deductive planning techniques are described for rule selection in question-answering systems. Rules typically represent inferential knowledge applicable to a given domain of discourse. Given a question-answering system containing a large number of such rules, a crucial problem arises in selecting those rules that are relevant and needed to answer particular input queries. A planning process has been implemented to find chains of rules that deductively infer the desired conclusions. The processes of forward chaining from assumptions and backward chaining from goals are combined to form middle-term chains which provide the basis of the planning mechanism. The applicability and generality of these techniques for use in other rule-based systems is also discussed. (Author)

Open PDF

Document Details

Document Type
Technical Report
Publication Date
May 01, 1977
Accession Number
ADA072092

Entities

People

  • Philip Klahr

Organizations

  • System Development Corporation

Tags

Communities of Interest

  • C4I

DTIC Thesaurus Topics

  • Artificial Intelligence
  • Artificial Intelligence Software
  • Computer Languages
  • Computer Programs
  • Computer Science
  • Computers
  • Consistency
  • Data Management
  • Databases
  • Demographic Cohorts
  • Language
  • Natural Languages
  • New York
  • Production
  • Rule Based Systems
  • Verification

Readers

  • Artificial Intelligence

Technology Areas

  • AI & ML
  • AI & ML - Information Retrieval