Knowledge Sources for Word-Level Translation Models

Philipp Koehn, Kevin Knight

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

We present various methods to train word-level translation models for statistical machine translation systems that use widely different knowledge sources ranging from parallel corpora and a bilingual lexicon to only monolingual corpora in two languages. Some novel methods are presented and previously published methods are reviewed. Also, a common evaluation metric enables the first quantitative comparison of these approaches.
























































Original languageEnglish
Title of host publicationProceedings of the 2001 Conference on Empirical Methods in Natural Language Processing
EditorsLillian Lee, Donna Harman
Pages27-35
Number of pages9
Publication statusPublished - 2001
Event2001 Conference on Empirical Methods in Natural Language Processing (EMNLP 2001) - McConomy Auditorium in University Center, Carnegie Mellon University, Pittsburgh, PA, United States
Duration: 3 Jun 20014 Jun 2001

Conference

Conference2001 Conference on Empirical Methods in Natural Language Processing (EMNLP 2001)
CountryUnited States
CityPittsburgh, PA
Period3/06/014/06/01

Cite this