Large-Scale Dictionary Construction for ForeignLanguage Tutoring and Interlingual Machine Translation

  • Authors:

  • Bonnie J. Dorr

  • Affiliations:

  • Department of Computer Science and UMIACS, University of Maryland, College Park, MD 20742, U.S.A. E-mail: bonnie@cs.umd.edu

  • Venue:
  • Machine Translation
  • Year:
  • 1998

Quantified Score

Hi-index 0.00

Visualization

Abstract

This paper describes techniques for automaticconstruction of dictionaries for use in large-scale foreign languagetutoring (FLT) and interlingual machine translation (MT) systems. Thedictionaries are based on a language-independent representation called“lexical conceptual structure” (LCS). A primary goal of the LCSresearch is to demonstrate that synonymous verb senses sharedistributional patterns. We show how thesyntax–semantics relation can be used to develop a lexical acquisitionapproach that contributes both toward the enrichment of existingonline resources and toward the development of lexicons containingmore complete information than is provided in any of these resourcesalone. We start by describing the structure of the LCS and showinghow this representation is used in FLT and MT. We then focus on theproblem of building LCS dictionaries for large-scale FLT and MT.First, we describe authoring tools for manual and semi-automaticconstruction of LCS dictionaries; we then present a more sophisticatedapproach that uses linguistic techniques for building word definitionsautomatically. These techniques have been implemented as part of aset of lexicon-development tools used in the milt FLT project.