Abstract
The research context of this paper is developing hybrid machine translation (MT) systems that exploit the advantages of linguistic rule-based and statistical MT systems. Arabic, as a morphologically rich language, is especially challenging even without addressing the hybridization question. In this paper, we describe the challenges in building an Arabic- English generation-heavy machine translation (GHMT) system and boosting it with statistical machine translation (SMT) components. We present an extensive evaluation of multiple system variants and report positive results on the advantages of hybridization.
| Original language | English (US) |
|---|---|
| Pages | 56-65 |
| Number of pages | 10 |
| State | Published - Dec 1 2006 |
| Event | 7th Biennial Conference of the Association for Machine Translation in the Americas, AMTA 2006 - Cambridge, MA, United States Duration: Aug 8 2006 → Aug 12 2006 |
Other
| Other | 7th Biennial Conference of the Association for Machine Translation in the Americas, AMTA 2006 |
|---|---|
| Country | United States |
| City | Cambridge, MA |
| Period | 8/8/06 → 8/12/06 |
Fingerprint
ASJC Scopus subject areas
- Language and Linguistics
- Human-Computer Interaction
- Software
Cite this
Challenges in building an Arabic-English GHMT system with SMT components. / Habash, Nizar; Dorr, Bonnie; Monz, Christof.
2006. 56-65 Paper presented at 7th Biennial Conference of the Association for Machine Translation in the Americas, AMTA 2006, Cambridge, MA, United States.Research output: Contribution to conference › Paper
}
TY - CONF
T1 - Challenges in building an Arabic-English GHMT system with SMT components
AU - Habash, Nizar
AU - Dorr, Bonnie
AU - Monz, Christof
PY - 2006/12/1
Y1 - 2006/12/1
N2 - The research context of this paper is developing hybrid machine translation (MT) systems that exploit the advantages of linguistic rule-based and statistical MT systems. Arabic, as a morphologically rich language, is especially challenging even without addressing the hybridization question. In this paper, we describe the challenges in building an Arabic- English generation-heavy machine translation (GHMT) system and boosting it with statistical machine translation (SMT) components. We present an extensive evaluation of multiple system variants and report positive results on the advantages of hybridization.
AB - The research context of this paper is developing hybrid machine translation (MT) systems that exploit the advantages of linguistic rule-based and statistical MT systems. Arabic, as a morphologically rich language, is especially challenging even without addressing the hybridization question. In this paper, we describe the challenges in building an Arabic- English generation-heavy machine translation (GHMT) system and boosting it with statistical machine translation (SMT) components. We present an extensive evaluation of multiple system variants and report positive results on the advantages of hybridization.
UR - http://www.scopus.com/inward/record.url?scp=71249102370&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=71249102370&partnerID=8YFLogxK
M3 - Paper
AN - SCOPUS:71249102370
SP - 56
EP - 65
ER -