A Paraphrase-Based Approach to Machine Translation Evaluation

Abstract

We propose a novel approach to automatic machine translation evaluation based on paraphrase identification. The quality of machine-generated output can be viewed as the extent to which the conveyed meaning matches the semantics of reference translations, independent of lexical and syntactic divergences. This idea is implemented ill linear regression models that attempt to capture human judgments of adequacy and fluency, based on features that have previously been shown to be effective for paraphrase identification. We evaluated our model using the output of three different MT systems from the 2004 NIST Arabic-to-English MT evaluation. Results show that models employing paraphrase-based features correlate better with human judgments than models based purely on existing automatic MT metrics.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Aug 01, 2005
Accession Number
ADA448032

Entities

People

  • Grazia Russo-lassner
  • Jimmy Lin
  • Philip Resnik

Organizations

  • University of Maryland

Tags

Communities of Interest

  • Autonomy

DTIC Thesaurus Topics

  • Applied Computer Science
  • Automated Text Summarization
  • Computational Linguistics
  • Computational Science
  • Computer Languages
  • Data Sets
  • Error Analysis
  • Information Science
  • Language
  • Linguistics
  • Machine Learning
  • Machine Translation
  • Natural Language Processing
  • Signal Processing
  • Supervised Machine Learning
  • Test And Evaluation
  • Translations

Fields of Study

  • Computer science

Readers

  • Computational Linguistics
  • Computational Modeling and Simulation

Technology Areas

  • AI & ML
  • AI & ML - Bayesian Inference
  • AI & ML - Machine Translation
  • AI & ML - Neural Networks