Dependency Graph-to-String Translation

Proceedings of the 2015 Conference on Empirical Methods in Natural Language Processing, pages 33–43, Lisbon, Portugal, 17-21 September 2015. c⃝2015 Association for Computational Linguistics. Dependency Graph-to-String Translation Liangyou Li Andy Way Qun Liu ADAPT Centre, School of Computing Dublin City University {liangyouli,away,qliu}@computing.dcu.ie Abstract Compared to tree grammars, graph gram- mars have stronger generative capacity over structures. Based on an edge re- placement grammar, in this paper we pro- pose to use a synchronous graph-to-string grammar for statistical machine transla- tion. The graph we use is directly con- verted from a dependency tree by labelling edges. We build our translation model in the log-linear framework with stan- dard features. Large-scale experiments on Chinese–English and German–English tasks show that our model is significantly better than the state-of-the-art hierarchical phrase-based (HPB) model and a recently improved dependency tree-to-string model on BLEU, METEOR and TER scores. Ex- periments also suggest that our model has better capability to perform long-distance reordering and is more suitable for trans- lating long sentences. 1 Introduction Compared to trees, which have dominated the field of natural language processing (NLP) for decades, graphs are more general for modelling natural lan- guages. The corresponding grammars for recog- nizing and producing graphs are more flexible and powerful than tree grammars. However, because of their high complexity, graph grammars have not been widely used in NLP. Recently, along with progress on graph-based meaning representation, hyperedge replacement grammars (HRG) (Drewes et al., 1997) have been revisited, explored and used for semantic-based machine translation (Jones et al., 2012). How- ever, the translation process is rather complex and the resources it relies on, namely abstract meaning corpora, are limited as well. As most available syntactic resources and tools are tree-based, in this paper we propose to con- vert dependency trees, which are usually taken as a kind of shallow semantic representation, to de- pendency graphs by labelling edges. We then use a synchronous version of edge replacement gram- mar (ERG) (Section 2), a special case of HRG, to translate these graphs. The resulting translation model has the same order of magnitude in terms of time complexity with the hierarchical phrase- based model (HPB) (Chiang, 2005) under a certain restriction (Section 3). Compared to dependency tree-to-string models, using ERG for graph-to-string translation brings some benefits (Section 3). Thanks to the stronger generative capacity of the grammar, our model can naturally translate siblings in a tree struc- ture, which are usually treated as non-syntactic phrases and handled by other techniques (Huck et al., 2014; Xie et al., 2014). Furthermore, com- pared to the known treelet approach (Quirk et al., 2005) and Dep2Str (Xie et al., 2011), our method not only uses treelets but also has a full capacity of reordering. We define our translation model (Section 4) in the log-linear framework (Och and Ney, 2002). Large-scale experiments (Section 5) on Chinese– English and German–English, two language pairs that have a high degree of syntactic reordering, show that our method significantly improves trans- lation quality over both HPB and …

Content truncated to 500 words