Text Analysis Using Automated Language Translators

Abstract

Text analysis is a new tool with many interesting possibilities for intelligence-gathering. Software being developed by Carnegie Mellon University can output a mental model of a text with the top six concepts in that text. This can be used to automatically analyze thousands of texts to search for keywords, find trends over time, or compare two different geographic areas. The problem is that most of the texts that intelligence analysts would use are not in English. Machine translators like the Forward Area Language Converter (FALCon) produce English text that is hard for the average person to read but this machine-translated text appears to be just as useful for text analysis as human-translated text. Using machines instead of humans to translate text can save intelligence agencies time and money.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Nov 01, 2006
Accession Number
ADA496372

Entities

People

  • Ian Mcculloh
  • John Stanford

Organizations

  • United States Military Academy

Tags

DTIC Thesaurus Topics

  • Abstracts
  • Forward Areas
  • Geographic Regions
  • Information Operations
  • Intelligence Analysts
  • Intelligence Collection
  • Language
  • Military Research
  • Translators

Fields of Study

  • Education

Readers

  • Computational Linguistics
  • Geospatial Intelligence and Artificial Intelligence Analytics

Technology Areas

  • AI & ML
  • AI & ML - Information Retrieval
  • AI & ML - Machine Translation