A Sense-Based Translation Model for Statistical Machine Translation

Proceedings of the 52nd Annual Meeting of the Association for Computational Linguistics, pages 1459–1469, Baltimore, Maryland, USA, June 23-25 2014. c⃝2014 Association for Computational Linguistics A Sense-Based Translation Model for Statistical Machine Translation Deyi Xiong and Min Zhang∗ Provincial Key Laboratory for Computer Information Processing Technology Soochow University, Suzhou, China 215006 {dyxiong, minzhang}@suda.edu.cn Abstract The sense in which a word is used deter- mines the translation of the word. In this paper, we propose a sense-based transla- tion model to integrate word senses into statistical machine translation. We build a broad-coverage sense tagger based on a nonparametric Bayesian topic model that automatically learns sense clusters for words in the source language. The pro- posed sense-based translation model en- ables the decoder to select appropriate translations for source words according to the inferred senses for these words us- ing maximum entropy classifiers. Our method is significantly different from pre- vious word sense disambiguation reformu- lated for machine translation in that the lat- ter neglects word senses in nature. We test the effectiveness of the proposed sense- based translation model on a large-scale Chinese-to-English translation task. Re- sults show that the proposed model sub- stantially outperforms not only the base- line but also the previous reformulated word sense disambiguation. 1 Introduction One of very common phenomena in language is that a plenty of words have multiple meanings. In the context of machine translation, such dif- ferent meanings normally produce different target translations. Therefore a natural assumption is that word sense disambiguation (WSD) may contribute to statistical machine translation (SMT) by provid- ing appropriate word senses for target translation selection with context features (Carpuat and Wu, 2005). ∗Corresponding author This assumption, however, has not been em- pirically verified in the early days. Carpuat and Wu (2005) adopt a standard formulation of WSD: predicting word senses that are defined on an ontology for ambiguous words. As they apply WSD to Chinese-to-English translation, they pre- dict word senses from a Chinese ontology HowNet and project the predicted senses to English glosses provided by HowNet. These glosses, used as the sense predictions of their WSD system, are inte- grated into a word-based SMT system either to substitute for translation candidates of their trans- lation model or to postedit the output of their SMT system. They report that WSD degenerates the translation quality of SMT. In contrast to the standard WSD formulation, Vickrey et al. (2005) reformulate the task of WSD for SMT as predicting possible target translations rather than senses for ambiguous source words. They show that such a reformulated WSD can im- prove the accuracy of a simplified word translation task. Following this WSD reformulation for SMT, Chan et al. (2007) integrate a state-of-the-art WSD system into a hierarchical phrase-based sys- tem (Chiang, 2005). Carpuat and Wu (2007) also use this reformulated WSD and further adapt it to multi-word phrasal disambiguation. They both re- port that the redefined WSD can significantly im- prove SMT. Although this reformulated WSD has proved helpful for SMT, one question …

Content truncated to 500 words