Applying Morphology Generation Models to Machine Translation

Proceedings of ACL-08: HLT, pages 514–522, Columbus, Ohio, USA, June 2008. c⃝2008 Association for Computational Linguistics Applying Morphology Generation Models to Machine Translation Kristina Toutanova Microsoft Research Redmond, WA, USA kristout@microsoft.com Hisami Suzuki Microsoft Research Redmond, WA, USA hisamis@microsoft.com Achim Ruopp Butler Hill Group Redmond, WA, USA v-acruop@microsoft.com Abstract We improve the quality of statistical machine translation (SMT) by applying models that predict word forms from their stems using extensive morphological and syntactic infor- mation from both the source and target lan- guages. Our inflection generation models are trained independently of the SMT system. We investigate different ways of combining the in- flection prediction component with the SMT system by training the base MT system on fully inflected forms or on word stems. We applied our inflection generation models in translating English into two morphologically complex languages, Russian and Arabic, and show that our model improves the quality of SMT over both phrasal and syntax-based SMT systems according to BLEU and human judge- ments. 1 Introduction One of the outstanding problems for further improv- ing machine translation (MT) systems is the diffi- culty of dividing the MT problem into sub-problems and tackling each sub-problem in isolation to im- prove the overall quality of MT. Evidence for this difficulty is the fact that there has been very little work investigating the use of such independent sub- components, though we started to see some success- ful cases in the literature, for example in word align- ment (Fraser and Marcu, 2007), target language cap- italization (Wang et al., 2006) and case marker gen- eration (Toutanova and Suzuki, 2007). This paper describes a successful attempt to in- tegrate a subcomponent for generating word inflec- tions into a statistical machine translation (SMT) system. Our research is built on previous work in the area of using morpho-syntactic information for improving SMT. Work in this area is motivated by two advantages offered by morphological analysis: (1) it provides linguistically motivated clustering of words and makes the data less sparse; (2) it cap- tures morphological constraints applicable on the target side, such as agreement phenomena. This sec- ond problem is very difficult to address with word- based translation systems, when the relevant mor- phological information in the target language is ei- ther non-existent or implicitly encoded in the source language. These two aspects of morphological pro- cessing have often been addressed separately: for example, morphological pre-processing of the input data is a common method of addressing the first as- pect, e.g. (Goldwater and McClosky, 2005), while the application of a target language model has al- most solely been responsible for addressing the sec- ond aspect. Minkov et al. (2007) introduced a way to address these problems by using a rich feature- based model, but did not apply the model to MT. In this paper, we integrate a model that predicts target word inflection in the translations of English into two morphologically complex languages (Rus- sian and Arabic) and show improvements in the MT output. We study several alternative …

Content truncated to 500 words