Representing Text Meaning for Multilingual Knowledge-Based Machine Translation.

Abstract

An interlingua capable of representing the meaning of natural language texts is a crucial component of a knowledge based machine translation system. In the Mikrokosmos project, researchers are defining the components of such an interlingua, or text meaning representation (TMR) language, through extensive analysis of Japanese and English texts in the domain of join business ventures. This paper describes the components of the TMR, providing examples of how certain phenomena are represented. The authors discuss their experience in analyzing the Japanese joint ventures corpus and its effect on TMR development. Dr. Ronald E. Dolan, a member of the Federal Research Division of the Library of Congress, is coauthor of the paper along with Lynn M. Carlson and Elizabeth L. Cooper of the U.S. Department of Defense, Fort George G. Meade, Maryland, and Steven J. Maiorano of the Office of Research and Development, Washington, D.C.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Oct 07, 1994
Accession Number
ADA302333

Entities

People

  • Elizabeth Cooper
  • Lynn Carlson
  • Ronald Dolan
  • Steve J. Maiorano

Organizations

  • Library of Congress

Tags

Communities of Interest

  • Materials and Manufacturing Processes

DTIC Thesaurus Topics

  • Classification
  • Cognitive Science
  • Commerce
  • Computational Linguistics
  • Congress
  • Department Of Defense
  • Japanese Language
  • Language
  • Linguistics
  • Machine Translation
  • Maryland
  • Natural Languages
  • New Jersey
  • Ontologies
  • Semantics
  • Standards
  • United States

Readers

  • Computational Linguistics
  • Defense Technology Research and Development.
  • Research Science/Academic Research

Technology Areas

  • AI & ML
  • AI & ML - Machine Translation