Estimating translation probabilities considering semantic recoverability of phrase retranslation

Hyoung Gyu Lee, Min Jeong Kim, Ying Xiu Quan, Hae-Chang Rim, So Young Park

Research output: Contribution to journalArticle

Abstract

The general method for estimating phrase translation probabilities consists of sequential processes: word alignment, phrase pair extraction, and phrase translation probability calculation. However, during this sequential process, errors may propagate from the word alignment step through the translation probability calculation step. In this paper, we propose a new method for estimating phrase translation probabilities that reduce the effects of error propagation. By considering the semantic recoverability of phrase retranslation, our method identifies incorrect phrase pairs that have propagated from alignment errors. Furthermore, we define retranslation similarity which represents the semantic recoverability of phrase retranslation, and use this when computing translation probabilities. Experimental results show that the proposed phrase translation estimation method effectively prevents a PBSMT system from selecting incorrect phrase pairs, and consistently improves the translation quality in various language pairs.

Original languageEnglish
Pages (from-to)897-901
Number of pages5
JournalIEICE Transactions on Information and Systems
VolumeE95-D
Issue number3
DOIs
Publication statusPublished - 2012 Mar 1

Fingerprint

Semantics

Keywords

  • Phrase retranslation
  • Phrase translation probability
  • Semantic recoverability
  • Statistical machine translation

ASJC Scopus subject areas

  • Electrical and Electronic Engineering
  • Software
  • Artificial Intelligence
  • Hardware and Architecture
  • Computer Vision and Pattern Recognition

Cite this

Estimating translation probabilities considering semantic recoverability of phrase retranslation. / Lee, Hyoung Gyu; Kim, Min Jeong; Quan, Ying Xiu; Rim, Hae-Chang; Park, So Young.

In: IEICE Transactions on Information and Systems, Vol. E95-D, No. 3, 01.03.2012, p. 897-901.

Research output: Contribution to journalArticle

Lee, Hyoung Gyu ; Kim, Min Jeong ; Quan, Ying Xiu ; Rim, Hae-Chang ; Park, So Young. / Estimating translation probabilities considering semantic recoverability of phrase retranslation. In: IEICE Transactions on Information and Systems. 2012 ; Vol. E95-D, No. 3. pp. 897-901.
@article{781183a5f9794b6d8a1a31f40e7e821b,
title = "Estimating translation probabilities considering semantic recoverability of phrase retranslation",
abstract = "The general method for estimating phrase translation probabilities consists of sequential processes: word alignment, phrase pair extraction, and phrase translation probability calculation. However, during this sequential process, errors may propagate from the word alignment step through the translation probability calculation step. In this paper, we propose a new method for estimating phrase translation probabilities that reduce the effects of error propagation. By considering the semantic recoverability of phrase retranslation, our method identifies incorrect phrase pairs that have propagated from alignment errors. Furthermore, we define retranslation similarity which represents the semantic recoverability of phrase retranslation, and use this when computing translation probabilities. Experimental results show that the proposed phrase translation estimation method effectively prevents a PBSMT system from selecting incorrect phrase pairs, and consistently improves the translation quality in various language pairs.",
keywords = "Phrase retranslation, Phrase translation probability, Semantic recoverability, Statistical machine translation",
author = "Lee, {Hyoung Gyu} and Kim, {Min Jeong} and Quan, {Ying Xiu} and Hae-Chang Rim and Park, {So Young}",
year = "2012",
month = "3",
day = "1",
doi = "10.1587/transinf.E95.D.897",
language = "English",
volume = "E95-D",
pages = "897--901",
journal = "IEICE Transactions on Information and Systems",
issn = "0916-8532",
publisher = "Maruzen Co., Ltd/Maruzen Kabushikikaisha",
number = "3",

}

TY - JOUR

T1 - Estimating translation probabilities considering semantic recoverability of phrase retranslation

AU - Lee, Hyoung Gyu

AU - Kim, Min Jeong

AU - Quan, Ying Xiu

AU - Rim, Hae-Chang

AU - Park, So Young

PY - 2012/3/1

Y1 - 2012/3/1

N2 - The general method for estimating phrase translation probabilities consists of sequential processes: word alignment, phrase pair extraction, and phrase translation probability calculation. However, during this sequential process, errors may propagate from the word alignment step through the translation probability calculation step. In this paper, we propose a new method for estimating phrase translation probabilities that reduce the effects of error propagation. By considering the semantic recoverability of phrase retranslation, our method identifies incorrect phrase pairs that have propagated from alignment errors. Furthermore, we define retranslation similarity which represents the semantic recoverability of phrase retranslation, and use this when computing translation probabilities. Experimental results show that the proposed phrase translation estimation method effectively prevents a PBSMT system from selecting incorrect phrase pairs, and consistently improves the translation quality in various language pairs.

AB - The general method for estimating phrase translation probabilities consists of sequential processes: word alignment, phrase pair extraction, and phrase translation probability calculation. However, during this sequential process, errors may propagate from the word alignment step through the translation probability calculation step. In this paper, we propose a new method for estimating phrase translation probabilities that reduce the effects of error propagation. By considering the semantic recoverability of phrase retranslation, our method identifies incorrect phrase pairs that have propagated from alignment errors. Furthermore, we define retranslation similarity which represents the semantic recoverability of phrase retranslation, and use this when computing translation probabilities. Experimental results show that the proposed phrase translation estimation method effectively prevents a PBSMT system from selecting incorrect phrase pairs, and consistently improves the translation quality in various language pairs.

KW - Phrase retranslation

KW - Phrase translation probability

KW - Semantic recoverability

KW - Statistical machine translation

UR - http://www.scopus.com/inward/record.url?scp=84863280489&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84863280489&partnerID=8YFLogxK

U2 - 10.1587/transinf.E95.D.897

DO - 10.1587/transinf.E95.D.897

M3 - Article

AN - SCOPUS:84863280489

VL - E95-D

SP - 897

EP - 901

JO - IEICE Transactions on Information and Systems

JF - IEICE Transactions on Information and Systems

SN - 0916-8532

IS - 3

ER -