Mova0.5 pre-release⌕
☰
home / docs / papers / Machine translation / Not an Interlingua, But Close: Comparison of English AMRs to Chinese and Czech

Not an Interlingua, But Close: Comparison of English AMRs to Chinese and Czech

What it is about

The authors take 100 English sentences that already have AMR graphs, translate them into Chinese and Czech, and write AMR graphs for the translations. The question is whether AMR could serve as a minimally divergent transfer layer in machine translation. The comparison is manual and qualitative. English–Chinese graphs are more compatible than English–Czech ones. For Czech, only 29 of the 100 sentences have structurally identical graphs (apart from argument labels), and 18 more differ only locally, so more than half differ in a deeper way. Causes include multiword expressions that are one concept in one language and several in another, and language-specific concept inventories that do not line up. The conclusion is that AMR is close to an interlingua for some language pairs but not for all, and its guidelines need refinement.

Ideas

For Mova

The code is tools/translate/meta.rs and measure.rs. I1 gives the Bible pairs a measure that is comparable to the paper's, and I2 addresses open problem 2 (phrases that carry function words). The paper's Chinese pair is closer than its Czech pair; the same pattern is worth checking for Mova's author pairs.

Topics: Machine translation

Cite

The paper itself, not this card. BibTeX from the library's record of the work.

@inproceedings{xue2014not,
  title = {{Not an Interlingua, But Close: Comparison of English AMRs to Chinese and Czech}},
  author = {Xue, Nianwen and Bojar, Ondřej and Hajič, Jan and Palmer, Martha and Urešová, Zdeňka and Zhang, Xiuhong},
  booktitle = {Proceedings of the Ninth International Conference on Language Resources and Evaluation (LREC'14)},
  year = {2014},
  url = {https://aclanthology.org/L14-1332/}
}

References

24 references, 8 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Banarescu, L., Bonial, C., Cai, S., Georgescu, M., Griffitt, K., Hermjakob, U., Knight, K., Koehn, P., Palmer, M., and Schneider, N. (2013). Abstract Meaning Representation for Sembanking. In Proceedings of the 7th Linguistic Annotation Workshop, Sophia, Bulgaria. find our card
  2. Bohnet, Bernd, Nivre, Joakim, Boguslavsky, Igor, Farkas, Richard, Ginter, Filip, and Hajič, Jan. (2013). Joint morphological and syntactic analysis for richly inflected languages. Transactions of the Association for Computational Linguistics, 1:415–428. find
  3. Charniak, Eugene. (2000). A maximum-entropy-inspired parser. In Proceedings of the 1st North American chapter of the Association for Computational Linguistics conference, pages 132–139. Association for Computational Linguistics. find
  4. Collins, Michael. (1999). Head-driven Statistical Models for Natural Language Parsing. Ph.D. thesis, University of Pennsylvania. find our card
  5. Copestake, Ann, Flickinger, Dan, Pollard, Carl, and Sag, IvanA. (2005). Minimal recursion semantics: An introduction. Research on Language and Computation, 3(2-3):281–332. find
  6. Gildea, Daniel and Jurafsky, Daniel. (2002). Automatic labeling of semantic roles. Computational linguistics, 28(3):245–288. find our card
  7. Hajič, Jan, Panevová, Jarmila, Urešová, Zdeňka, Bémová, Alevtina, Kolářová, Veronika, and Pajas, Petr. (2003). PDT-VALLEX: Creating a large-coverage valency lexicon for treebank annotation. In Nivre, Joakim and Hinrichs, Erhard, editors, Proceedings of The Second Workshop on Treebanks and Linguistic Theories, volume 9 of Mathematical Modeling in Physics, Engineering and Cognitive Sciences, pages 57–68, Vaxjo, Sweden. Vaxjo University Press. find
  8. Hajič, Jan, Ciaramita, Massimiliano, Johansson, Richard, Kawahara, Daisuke, Martı́, Maria, Màrquez, Lluı́s, Meyers, Adam, Nivre, Joakim, Padó, Sebastian, Štěpánek, Jan, Straňák, Pavel, Surdeanu, Mihai, Xue, Nianwen, and Zhang, Yi. (2009). The CoNLL-2009 Shared Task: Syntactic and Semantic Dependencies in Multiple Languages. In Hajič, Jan, editor, Proceedings of the Thirteenth Conference on Computational Natural Language Learning (CoNLL): Shared Task, pages 1–18, Boulder, CO, USA. Association for Computational Linguistics. find our card
  9. Hajič, Jan, Hajičová, Eva, Panevová, Jarmila, Sgall, Petr, Bojar, Ondřej, Cinková, Silvie, Fučı́ková, Eva, Mikulová, Marie, Pajas, Petr, Popelka, Jan, Semecký, Jiřı́, Šindlerová, Jana, Štěpánek, Jan, Toman, Josef, Urešová, Zdeňka, and Žabokrtský, Zdeněk. (2012). Announcing Prague Czech-English Dependency Treebank 2.0. In Proceedings of the Eighth International Language Resources and Evaluation Conference (LREC’12), pages 3153–3160, Istanbul, Turkey, May. ELRA, European Language Resources Association. find
  10. Hajič, Jan, Böhmová, Alena, Hajicová, Eva, and Hladká, Barbora. (2003). The Prague Dependency Treebank: A Three Level Annotation Scenario. In Abeillé, Anne, editor, Treebanks: Building and Using Annotated Corpora. Kluwer Academic Publishers. find
  11. Jones, Bevan, Andreas, Jacob, Bauer, Daniel, Hermann, Karl Moritz, and Knight, Kevin. (2012). Semantics-based machine translation with hyperedge replacement grammars. In COLING, pages 1359–1376. find our card
  12. Marcus, Mitchell P., Santorini, Beatrice, and Marcinkiewicz, Mary Ann. (1993). Building a large annotated corpus of english: The penn treebank. Computational Linguistics, 19(2):313–330. find our card
  13. Meyers, A., Reeves, R., Macleod, C., Szekely, R., Zielinska, V., Young, B., and Grishman, R. (2004). The Nom-Bank Project: An Interim Report. In Proceedings of the NAACL/HLT Workshop on Frontiers in Corpus Annotation, pages 24–31, Boston, Massachusetts. find
  14. Palmer, Martha, Gildea, Daniel, and Kingsbury, Paul. (2005). The Proposition Bank: An annotated corpus of semantic roles. Computational Linguistics, 31(1):71– 106. find our card
  15. Petrov, Slav and Klein, Dan. (2007). Improved inference for unlexicalized parsing. In HLT-NAACL, pages 404– 411. find
  16. Pradhan, Sameer S, Ward, Wayne, Hacioglu, Kadri, Martin, James H, and Jurafsky, Daniel. (2004). Shallow semantic parsing using support vector machines. In HLT-NAACL, pages 233–240. find
  17. Prasad, Rashmi, Dinesh, Nikhil, Lee, Alan, Miltsakaki, Eleni, Robaldo, Livio, Joshi, Aravind, and Webber, Bonnie. (2008). The Penn Discourse Treebank 2.0. In Proceedings of the 6th International Conference on Language Resources and Evaluation (LREC 2008), Marrakech, Morocco. find our card
  18. Pustejovsky, James, Hanks, Patrick, Saurı́, Roser, See, Andrew, Day, David, Ferro, Lisa, Gaizauskas, Robert, Lazo, Marcia, Setzer, Andrea, and Sundheim, Beth. (2003). The TimeBank Corpus. Corpus Linguistics, pages 647– 656. find
  19. Saurı́, Roser and Pustejovsky, James. (2009). Factbank: a corpus annotated with event factuality. Language resources and evaluation, 43(3):227–268. find
  20. Sgall, Petr, Hajičová, Eva, and Panevová, Jarmila. (1986). The Meaning of the Sentence and Its Semantic and Pragmatic Aspects. Academia/Reidel Publishing Company, Prague, Czech Republic/Dordrecht, Netherlands. find
  21. Srikumar, Vivek and Roth, Dan. (2013). Modeling semantic relations expressed by prepositions. Transactions of the Association for Computational Linguistics, 1:231– 242. find
  22. Urešová, Zdeňka. (2011). Valenčnı́ slovnı́k Pražského závislostnı́ho korpusu (PDT-Vallex). Studies in Computational and Theoretical Linguistics. Ústav formálnı́ a aplikované lingvistiky, Praha, Czechia. find
  23. Xue, Nianwen and Palmer, Martha. (2004). Calibrating features for semantic role labeling. In EMNLP, pages 88–94. find
  24. Xue, Nianwen, Xia, Fei, dong Chiou, Fu, and Palmer, Martha. (2005). The Penn Chinese TreeBank: Phrase Structure Annotation of a Large Corpus. Natural Language Engineering, 11(2):207–238. find

This page in Markdown → https://md.mova.elamur.ai/free/en/docs/papers/xue-2014-not-interlingua-but-close-comparison/