2009 AcquiringTranslEquivOfMWEbyNormCorrFreq

• (Bai et al., 2009) ⇒ Ming-Hong Bai, Jia-Ming You, Keh-Jiann Chen, Jason S. Chang. (2009). “Acquiring Translation Equivalences of Multiword Expressions by Normalized Correlation Frequencies.” In: Proceedings of EMNLP Conference (EMNLP 2009)

Subject Headings: Automated Natural Language Translation Task, Multiword Expression.

Cited By

• Adam Lopez. http://nlpers.blogspot.com/2009/08/acs-machine-translation-papers-at-emnlp.html
  • Bai et al. focus on the problem of acquiring multiword expressions (i.e. idioms), showing why typical word alignment methods fail, and using a combination of statistical association measures and heuristics to fix the problem, with small gains in Chinese-English.

Quotes

Abstract

• In this paper, we present an algorithm for extracting translations of any given multiword expression from parallel corpora. Given a multiword expression to be translated, the method involves extracting a short list of target candidate words from parallel corpora based on scores of normalized frequency, generating possible translations and filtering out common subsequences, and selecting the top-n possible translations using the Dice coefficient. Experiments show that our approach outperforms the word alignmentbased and other naive association-based methods. We also demonstrate that adopting the extracted translations can significantly improve the performance of the Moses machine translation system.

1 Introduction

• Translation of multiword expressions (MWEs), such as compound words, phrases, collocations and idioms, is important for many NLP tasks, including the techniques are helpful for dictionary compilation, cross language information retrieval, second language learning, and machine translation. (Smadja et al., 1996; Gao et al., 2002; Wu and Zhou, 2003). However, extracting exact translations of MWEs is still an open problem, possibly because the senses of many MWEs are not compositional (Yamamoto and Matsumoto, 2000), i.e., their translations are not compositions of the translations of individual words. For example, the Chinese idiom 坐視不理 should be translated as “turn a blind eye,” which has no direct relation with respect to the translation of each constituent (i.e., “to sit”, “to see” and “to ignore”) at the word level.
• Previous SMT systems (e.g., Brown et al., 1993) used a word-based translation model which assumes that a sentence can be translated into other languages by translating each word into one or more words in the target language. Since many concepts are expressed by idiomatic multiword expressions instead of single words, and different languages may realize the same concept using different numbers of words (Ma et al., 2007; Wu, 1997), word alignment based methods, which are highly dependent on the probability information at the lexical level, are not well suited for this type of translation.
• To address the above problem, some methods have been proposed for extending word alignments to phrase alignments. For example, Och et al. (1999) proposed the so-called grow-diag--final heuristic method for extending word alignments to phrase alignments. The method is widely used and has achieved good results for phrase-based statistical machine translation. (Och et al., 1999; Koehn et al., 2003; Liang et al., 2006). Instead of using heuristic rules, Ma et al. (2008) showed that syntactic information, e.g., phrase or dependency structures, is useful in extending the word-level alignment. However, the above methods still depend on word-based alignment models, so they are not well suited to extracting the translation equivalences of semantically opaque MWEs due to the lack of word-level relations between the translational correspondences. Moreover, the aligned phrases are not precise enough to be used in many NLP applications like dictionary compilation, which require high quality translations. Association-based methods, e.g., the Dice coefficient, are widely used to extract translations of MWEs. (Kupiec, 1993; Smadja et al., 1996; Kitamura and Matsumoto, 1996; Yamamoto and Matsumoto, 2000; Melamed, 2001).
• The advantage of such methods is that association relations are established at the phrase level instead of the lexical level, so they have the potential to resolve the above-mentioned translation problem. However, when applying association-based methods, we have to consider the following complications. The first complication, which we call the contextual effect, causes the extracted translation to contain noisy words. For,

	Author	volume	Date Value	title	type	journal	titleUrl	doi	note	year
2009 AcquiringTranslEquivOfMWEbyNormCorrFreq	Ming-Hong Bai Jia-Ming You Keh-Jiann Chen Jason S. Chang			Acquiring Translation Equivalences of Multiword Expressions by Normalized Correlation Frequencies		Proceedings of EMNLP Conference	http://aclweb.org/anthology-new/D/D09/D09-1050.pdf			2009