Project-specific Machine Translation

Abstract

This report details a project-specific method for translating a book-length text. The method used a human-in-the-loop strategy to iteratively and incrementally produce a progressively more accurate statistical machine translation system. The system generated rough-draft translations of chunks of text that were post-edited by an expert translator. The method was used to produce a high-quality translation of the Army Field Manual (FM) 7-8 Infantry Rifle Platoon and Squad from English into the Afghan language Pashto. The results show that through a process of over-fitting, this method leads to systems that are optimized for aiding a single translator.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Dec 01, 2011
Accession Number
ADA554967

Entities

People

  • John J. Morgan

Organizations

  • United States Army Research Laboratory

Tags

Communities of Interest

  • Human Systems
  • Materials and Manufacturing Processes
  • Weapons Technologies

DTIC Thesaurus Topics

  • Armored Vehicles
  • Army
  • Books
  • Computational Science
  • Handbooks
  • Infantry
  • Information Science
  • Language
  • Linguistics
  • Machine Translation
  • Machines
  • Military Research
  • Natural Language Processing
  • Natural Languages
  • Pashto Language
  • Translations
  • Translators

Readers

  • Computational Linguistics
  • Military Training and Readiness Simulation
  • Team-Based Human-Centered Cognitive Task Decision Making and Information Performance.

Technology Areas

  • AI & ML
  • AI & ML - Machine Translation