Fast translation rule matching for syntax-based statistical machine translation

Hui Zhang; Min Zhang; Haizhou Li; Chew Lim Tan

doi:10.3115/1699571.1699647

Fast translation rule matching for syntax-based statistical machine translation

Hui Zhang^*
, Min Zhang
, Haizhou Li
, Chew Lim Tan

^*Corresponding author for this work

Research output: Contribution to conference › Paper › peer-review

9 Scopus citations

Abstract

In a linguistically-motivated syntax-based translation system, the entire translation process is normally carried out in two steps, translation rule matching and target sentence decoding using the matched rules. Both steps are very time-consuming due to the tremendous number of translation rules, the exhaustive search in translation rule matching and the complex nature of the translation task itself. In this paper, we propose a hyper-tree-based fast algorithm for translation rule matching. Experimental results on the NIST MT-2003 Chinese-English translation task show that our algorithm is at least 19 times faster in rule matching and is able to help to save 57% of overall translation time over previous methods when using large fragment translation rules.

Original language	English
Pages	1037-1045
Number of pages	9
DOIs	https://doi.org/10.3115/1699571.1699647
State	Published - 2009
Externally published	Yes
Event	2009 Conference on Empirical Methods in Natural Language Processing, EMNLP 2009, Held in Conjunction with ACL-IJCNLP 2009 - Singapore, Singapore Duration: 6 Aug 2009 → 7 Aug 2009

Conference

Conference	2009 Conference on Empirical Methods in Natural Language Processing, EMNLP 2009, Held in Conjunction with ACL-IJCNLP 2009
Country/Territory	Singapore
City	Singapore
Period	6/08/09 → 7/08/09

Access to Document

10.3115/1699571.1699647

Cite this

@conference{4ad808daf2cc437da820c9dc486ad469,

title = "Fast translation rule matching for syntax-based statistical machine translation",

abstract = "In a linguistically-motivated syntax-based translation system, the entire translation process is normally carried out in two steps, translation rule matching and target sentence decoding using the matched rules. Both steps are very time-consuming due to the tremendous number of translation rules, the exhaustive search in translation rule matching and the complex nature of the translation task itself. In this paper, we propose a hyper-tree-based fast algorithm for translation rule matching. Experimental results on the NIST MT-2003 Chinese-English translation task show that our algorithm is at least 19 times faster in rule matching and is able to help to save 57\% of overall translation time over previous methods when using large fragment translation rules.",

author = "Hui Zhang and Min Zhang and Haizhou Li and Tan, \{Chew Lim\}",

year = "2009",

doi = "10.3115/1699571.1699647",

language = "英语",

pages = "1037--1045",

note = "2009 Conference on Empirical Methods in Natural Language Processing, EMNLP 2009, Held in Conjunction with ACL-IJCNLP 2009 ; Conference date: 06-08-2009 Through 07-08-2009",

}

Fast translation rule matching for syntax-based statistical machine translation. / Zhang, Hui; Zhang, Min; Li, Haizhou et al.
2009. 1037-1045 Paper presented at 2009 Conference on Empirical Methods in Natural Language Processing, EMNLP 2009, Held in Conjunction with ACL-IJCNLP 2009, Singapore, Singapore.

Research output: Contribution to conference › Paper › peer-review

TY - CONF

T1 - Fast translation rule matching for syntax-based statistical machine translation

AU - Zhang, Hui

AU - Zhang, Min

AU - Li, Haizhou

AU - Tan, Chew Lim

PY - 2009

Y1 - 2009

N2 - In a linguistically-motivated syntax-based translation system, the entire translation process is normally carried out in two steps, translation rule matching and target sentence decoding using the matched rules. Both steps are very time-consuming due to the tremendous number of translation rules, the exhaustive search in translation rule matching and the complex nature of the translation task itself. In this paper, we propose a hyper-tree-based fast algorithm for translation rule matching. Experimental results on the NIST MT-2003 Chinese-English translation task show that our algorithm is at least 19 times faster in rule matching and is able to help to save 57% of overall translation time over previous methods when using large fragment translation rules.

AB - In a linguistically-motivated syntax-based translation system, the entire translation process is normally carried out in two steps, translation rule matching and target sentence decoding using the matched rules. Both steps are very time-consuming due to the tremendous number of translation rules, the exhaustive search in translation rule matching and the complex nature of the translation task itself. In this paper, we propose a hyper-tree-based fast algorithm for translation rule matching. Experimental results on the NIST MT-2003 Chinese-English translation task show that our algorithm is at least 19 times faster in rule matching and is able to help to save 57% of overall translation time over previous methods when using large fragment translation rules.

UR - https://www.scopus.com/pages/publications/80053243865

U2 - 10.3115/1699571.1699647

DO - 10.3115/1699571.1699647

M3 - 论文

AN - SCOPUS:80053243865

SP - 1037

EP - 1045

T2 - 2009 Conference on Empirical Methods in Natural Language Processing, EMNLP 2009, Held in Conjunction with ACL-IJCNLP 2009

Y2 - 6 August 2009 through 7 August 2009

ER -

Fast translation rule matching for syntax-based statistical machine translation

Abstract

Conference

Access to Document

Other files and links

Fingerprint

Cite this