The use of a structural n-gram language model in generation-heavy hybrid machine translation

Research output: Contribution to journalArticle

Abstract

This paper describes the use of a statistical structural N-gram model in the natural language generation component of a Spanish-English generation-heavy hybrid machine translation system. A structural N-gram model captures the relationship between words in a dependency representation without taking into account the overall structure at the phrase level. The model is used together with other components in the system for lexical and structural selection. An evaluation of the machine translation system shows that the use of structural N-grams decreases runtime by 60% with no loss in translation quality.

Original languageEnglish (US)
Pages (from-to)61-69
Number of pages9
JournalLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume3123
StatePublished - Dec 1 2004

Fingerprint

Machine Translation
N-gram
Language Model
Natural Language Generation
Model
Decrease
Evaluation

ASJC Scopus subject areas

  • Theoretical Computer Science
  • Computer Science(all)

Cite this

@article{c338b5993b8443e38bd8802761251543,
title = "The use of a structural n-gram language model in generation-heavy hybrid machine translation",
abstract = "This paper describes the use of a statistical structural N-gram model in the natural language generation component of a Spanish-English generation-heavy hybrid machine translation system. A structural N-gram model captures the relationship between words in a dependency representation without taking into account the overall structure at the phrase level. The model is used together with other components in the system for lexical and structural selection. An evaluation of the machine translation system shows that the use of structural N-grams decreases runtime by 60{\%} with no loss in translation quality.",
author = "Nizar Habash",
year = "2004",
month = "12",
day = "1",
language = "English (US)",
volume = "3123",
pages = "61--69",
journal = "Lecture Notes in Computer Science",
issn = "0302-9743",
publisher = "Springer Verlag",

}

TY - JOUR

T1 - The use of a structural n-gram language model in generation-heavy hybrid machine translation

AU - Habash, Nizar

PY - 2004/12/1

Y1 - 2004/12/1

N2 - This paper describes the use of a statistical structural N-gram model in the natural language generation component of a Spanish-English generation-heavy hybrid machine translation system. A structural N-gram model captures the relationship between words in a dependency representation without taking into account the overall structure at the phrase level. The model is used together with other components in the system for lexical and structural selection. An evaluation of the machine translation system shows that the use of structural N-grams decreases runtime by 60% with no loss in translation quality.

AB - This paper describes the use of a statistical structural N-gram model in the natural language generation component of a Spanish-English generation-heavy hybrid machine translation system. A structural N-gram model captures the relationship between words in a dependency representation without taking into account the overall structure at the phrase level. The model is used together with other components in the system for lexical and structural selection. An evaluation of the machine translation system shows that the use of structural N-grams decreases runtime by 60% with no loss in translation quality.

UR - http://www.scopus.com/inward/record.url?scp=35048826742&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=35048826742&partnerID=8YFLogxK

M3 - Article

AN - SCOPUS:35048826742

VL - 3123

SP - 61

EP - 69

JO - Lecture Notes in Computer Science

JF - Lecture Notes in Computer Science

SN - 0302-9743

ER -