Machine translation between Hebrew and Arabic

Reshef Shilon; Nizar Habash; Alon Lavie; Shuly Wintner

doi:10.1007/s10590-011-9103-z

Machine translation between Hebrew and Arabic

Reshef Shilon, Nizar Habash, Alon Lavie, Shuly Wintner

Department of Computer Science

Research output: Contribution to journal › Article › peer-review

Abstract

Hebrew and Arabic are related but mutually incomprehensible languages with complex morphology and scarce parallel corpora. Machine translation between the two languages is therefore interesting and challenging. We discuss similarities and differences between Hebrew and Arabic, the benefits and challenges that they induce, respectively, and their implications on machine translation. We highlight the shortcomings of using English as a pivot language and advocate a direct, transfer-based and linguistically-informed (but still statistical, and hence scalable) approach. We report preliminary results of the two systems we are currently developing, for translation in both directions.

Original language	English
Pages (from-to)	177-195
Number of pages	19
Journal	Machine Translation
Volume	26
Issue number	1-2
DOIs	https://doi.org/10.1007/s10590-011-9103-z
State	Published - Mar 2012

Bibliographical note

Funding Information:
Acknowledgements We are grateful to Gennadi Lembersky for his help. Reshef Shilon and Shuly Wint-ner were partly supported by The Israel Science Foundation (grant No. 137/06). Alon Lavie’s work was supported in part by NSF grants IIS-0534217 and IIS-0915327. All of Nizar Habash’s contributions were funded by Columbia University’s Center for Computational Learning Systems (CCLS).

Keywords

Arabic
Hebrew
Transfer-based MT

ASJC Scopus subject areas

Software
Language and Linguistics
Linguistics and Language
Artificial Intelligence

Access to Document

10.1007/s10590-011-9103-z

Cite this

@article{a6ba1cabc29a4015a918938d5338f4f5,

title = "Machine translation between Hebrew and Arabic",

abstract = "Hebrew and Arabic are related but mutually incomprehensible languages with complex morphology and scarce parallel corpora. Machine translation between the two languages is therefore interesting and challenging. We discuss similarities and differences between Hebrew and Arabic, the benefits and challenges that they induce, respectively, and their implications on machine translation. We highlight the shortcomings of using English as a pivot language and advocate a direct, transfer-based and linguistically-informed (but still statistical, and hence scalable) approach. We report preliminary results of the two systems we are currently developing, for translation in both directions.",

keywords = "Arabic, Hebrew, Transfer-based MT",

author = "Reshef Shilon and Nizar Habash and Alon Lavie and Shuly Wintner",

note = "Funding Information: Acknowledgements We are grateful to Gennadi Lembersky for his help. Reshef Shilon and Shuly Wint-ner were partly supported by The Israel Science Foundation (grant No. 137/06). Alon Lavie{\textquoteright}s work was supported in part by NSF grants IIS-0534217 and IIS-0915327. All of Nizar Habash{\textquoteright}s contributions were funded by Columbia University{\textquoteright}s Center for Computational Learning Systems (CCLS).",

year = "2012",

month = mar,

doi = "10.1007/s10590-011-9103-z",

language = "English",

volume = "26",

pages = "177--195",

journal = "Machine Translation",

issn = "0922-6567",

publisher = "Springer Netherlands",

number = "1-2",

}

TY - JOUR

T1 - Machine translation between Hebrew and Arabic

AU - Shilon, Reshef

AU - Habash, Nizar

AU - Lavie, Alon

AU - Wintner, Shuly

N1 - Funding Information: Acknowledgements We are grateful to Gennadi Lembersky for his help. Reshef Shilon and Shuly Wint-ner were partly supported by The Israel Science Foundation (grant No. 137/06). Alon Lavie’s work was supported in part by NSF grants IIS-0534217 and IIS-0915327. All of Nizar Habash’s contributions were funded by Columbia University’s Center for Computational Learning Systems (CCLS).

PY - 2012/3

Y1 - 2012/3

N2 - Hebrew and Arabic are related but mutually incomprehensible languages with complex morphology and scarce parallel corpora. Machine translation between the two languages is therefore interesting and challenging. We discuss similarities and differences between Hebrew and Arabic, the benefits and challenges that they induce, respectively, and their implications on machine translation. We highlight the shortcomings of using English as a pivot language and advocate a direct, transfer-based and linguistically-informed (but still statistical, and hence scalable) approach. We report preliminary results of the two systems we are currently developing, for translation in both directions.

AB - Hebrew and Arabic are related but mutually incomprehensible languages with complex morphology and scarce parallel corpora. Machine translation between the two languages is therefore interesting and challenging. We discuss similarities and differences between Hebrew and Arabic, the benefits and challenges that they induce, respectively, and their implications on machine translation. We highlight the shortcomings of using English as a pivot language and advocate a direct, transfer-based and linguistically-informed (but still statistical, and hence scalable) approach. We report preliminary results of the two systems we are currently developing, for translation in both directions.

KW - Arabic

KW - Hebrew

KW - Transfer-based MT

UR - http://www.scopus.com/inward/record.url?scp=84857710778&partnerID=8YFLogxK

U2 - 10.1007/s10590-011-9103-z

DO - 10.1007/s10590-011-9103-z

M3 - Article

AN - SCOPUS:84857710778

SN - 0922-6567

VL - 26

SP - 177

EP - 195

JO - Machine Translation

JF - Machine Translation

IS - 1-2

ER -

Machine translation between Hebrew and Arabic

Abstract

Bibliographical note

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this