Abstract
Many research works have reported discontinuous phrase translation can significantly improve translation quality, but experiments are conducted in only one translation direction (e.g. Chinese-to-English) with only one language pair. Thus, two questions remain that whether the discontinuous rules are always much helpful in different language pairs? Furthermore, what kind of discontinuous rules (e.g. source-side discontinuity or target-side discontinuity) contributes most to the performance improvement? To answer these two questions, this paper conducts a comparative study on the contribution of different kinds of discontinuous rules in both translation directions with various language pairs. Then, with this comparative study, this paper proposes a role-based rule filtering strategy to filter the large amount of discontinuous rules that contribute very little to translation quality.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Bod, R.: Unsupervised syntax-based machine translation: The contribution of discontiguous phrases. In: Proc. of MT Summit 2007, pp. 51–57 (2007)
Chiang, D.: Hierarchical phrase-based translation. Computational Linguistics 33(2), 201–228 (2007)
Gally, M., Graehl, J., Knight, K., Marcu, D., DeNeefe, S., Wang, W., Thayer, I.: Scalable inference and training of context-rich syntactic translation models. In: Proc. of ACL-COLING 2006 (2006)
Galley, M., Manning, C.D.: Accurate Non-Hierarchical Phrase-Based Translation. In: Proc. of NAACL 2010 (2010)
Iglesias, G., Gispert, A., Banga, E.R., Byrne, W.: Rule filtering by pattern for efficient hierarchical translation. In: Proc. of EACL 2009, pp. 380–388 (2009)
Li, Z., Callison-Burch, C., Dyer, C., Ganitkevitch, J., Khudanpur, S., Schwartz, L., Thornton, W.N.G., Weese, J., Zaidan, O.F.: Joshua: An open source toolkit for parsing-based machine translation. In: Proc. of ACL 2009, pp. 135–139 (2009)
Lopez, A.: Machine translation by pattern matching. Ph.D. Thesis, University of Maryland at College Park (2007)
Quirk, C., Menezes, A.: Do we need phrases? Challenging the conventional wisdom in statistical machine translation. In: Proc. of HLT-NAACL 2006, pp. 9–16 (2006)
Shen, L., Xu, J., Weischedel, R.: A new string-to-dependency machine translation algorithm with a target dependency language model. In: Proc. of ACL 2008: HLT, pp. 577–585 (2008)
Simard, M., Cancedda, N., Cavestro, B., Dymetman, M., Gaussier, E., Goutte, C., Yamada, K., Langlais, P., Mauser, A.: Translating with non-contiguous phrases. In: Proc. of HLT-EMNLP 2005 (2005)
Sun, J., Zhang, M., Tan, C.L.: A non-contiguous tree sequence alignment-based model for statistical machine translation. In: Proc. of ACL-IJCNLP 2009, pp. 914–922 (2009)
Zollmann, A., Venugopal, A., Och, F., Ponte, J.: A systematic comparison of phrase-based, hierarchical and syntax-augmented statistical MT. In: Proc. of COLING 2008, pp. 1145–1152 (2008)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2012 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Zhang, J., Zong, C. (2012). A Comparative Study on Discontinuous Phrase Translation. In: Zhou, M., Zhou, G., Zhao, D., Liu, Q., Zou, L. (eds) Natural Language Processing and Chinese Computing. NLPCC 2012. Communications in Computer and Information Science, vol 333. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-34456-5_16
Download citation
DOI: https://doi.org/10.1007/978-3-642-34456-5_16
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-642-34455-8
Online ISBN: 978-3-642-34456-5
eBook Packages: Computer ScienceComputer Science (R0)