Machine Translation Based on Constraint-Based Synchronous Grammar

R. Dale et al. (Eds.): IJCNLP 2005, LNAI 3651, pp. 612 – 623, 2005. © Springer-Verlag Berlin Heidelberg 2005 Machine Translation Based on Constraint-Based Synchronous Grammar Fai Wong1, Dong-Cheng Hu1, Yu-Hang Mao1, Ming-Chui Dong2, and Yi-Ping Li2 1 Speech and Language Processing Research Center, Department of Automation, Tsinghua University, 100084 Beijing huangh01@mails.tsinghua.edu.cn {hudc, myh-dau}@mail.tsinghua.edu.cn 2 Faculty of Science and Technology of University of Macao, Av. Padre Tomás Pereira S.J., Taipa, Macao {dmc, ypli}@umac.mo Abstract. This paper proposes a variation of synchronous grammar based on the formalism of context-free grammar by generalizing the first component of productions that models the source text, named Constraint-based Synchronous Grammar (CSG). Unlike other synchronous grammars, CSG allows multiple target productions to be associated to a single source production rule, which can be used to guide a parser to infer different possible translational equivalences for a recognized input string according to the feature constraints of symbols in the pattern. Furthermore, CSG is augmented with independent rewriting that al- lows expressing discontinuous constituents in the inference rules. It turns out that such grammar is more expressive to model the translational equivalences of parallel texts for machine translation, and in this paper, we propose the use of CSG as a basis for building a machine translation (MT) system for Portuguese to Chinese translation. 1 Introduction In machine translation, to analyze the structure deviations of languages pair hence to carry out the transformation from one language into another as the target translation is the kernel part in a translation system, and this requires a large amount of structural transformations in both grammatical and concept level. The problems of syntactic com- plexity and word sense ambiguity have been the major obstacles to produce promising quality of translation. In order to overcome the obstacles and hence to improve the qual- ity of translation systems, several alternative approaches have been proposed. As stated in [1], much of the theoretical linguistics can be formulated in a very natural manner as stating correspondences between layers of representations. In simi- lar, many problems in natural language processing, in particular language translation and grammar rewriting systems, can be expressed as transduction through the use of synchronous formalisms [2,3,4,5,6]. Recently, synchronous grammars are becoming more and more popular for the formal description of parallel texts representing trans- lations for the same document. The underlying idea of such formalisms is to combine two generative devices through a pairing of their productions in such a way that right Machine Translation Based on Constraint-Based Synchronous Grammar 613 hand side non-terminal symbols in the paired productions are linked. However, such formalisms are less expressive and unable to express mutual translations that have different lengths and crossing dependencies. Moreover, synchronous formalisms do not deal with unification and feature structures, as in unification-based formalisms, that give patterns additional power for describing constraints on features. For exam- ples, Multiple Context-Free Grammar [4], where functions are engaged to the non- terminal symbols in the productions to further interpreting the symbols in target gen- eration. In …

Content truncated to 500 words