Elamur Research⌕
☰
home / papers / 2019 / Acquiring Annotated Data with Cross-lingual Explicitation for Implicit Discourse Relation Classification

Acquiring Annotated Data with Cross-lingual Explicitation for Implicit Discourse Relation Classification

Summary

Implicit discourse relations, which lack an explicit connective, are hard to classify, and the PDTB offers only about 18k annotated training instances. This work builds on Shi et al. (2017), who automatically labeled English implicit relations by finding cases where translators added a connective. The authors ask whether the choice of target language matters and whether combining several languages yields cleaner labels. They rebuild Europarl into document-aligned English–French–German–Czech text, so that adjacent sentences remain coherent. They then back-translate each language into English with Moses SMT and run a PDTB-style parser to find connectives that have become explicit. A majority vote across the three languages assigns labels with one, two or three agreeing votes. Training on instances where at least two back-translations agree gave the best second-level PDTB classification. It significantly beat both a model without extra data and models using a single language. The study shows that multilingual agreement can filter noise from automatically acquired discourse data. It also helps uncover new discourse signals and relations that hold at the same time.

Key points

  • Europarl input of about 2.05M En–Fr, 1.96M En–De and 0.65M En–Cz pairs was reduced to 532,542 four-way parallel sentence pairs in 6,105 documents.
  • Three Moses SMT back-translation systems were used, with training-set BLEU of 66.20 (French), 65.30 (German) and 69.05 (Czech). SMT was preferred over NMT because it translates connectives more stably.
  • Explicit connectives in the back-translations were detected with the Lin et al. (2014) PDTB-style parser, which has about 96% accuracy on connective identification. Only explicit relations with adjacent arguments were kept.
  • Evaluation used the conventional PDTB second-level 11-way setting, including cross-validation. The Two Votes subset produced the most useful additional data.
  • German back-translations contained proportionally more explicitated conjunction relations. Qualitative analysis showed that consistent translations reveal non-PDTB cues (e.g., "what is more") and ambiguous or co-occurring relations.

Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.

Fields: Linguistics & languages

Topics: Discourse analysis · Machine translation

Cited by

6 papers of our library cite it (of the 140673 whose bibliographies we read); 5 of them have a page here, 1 not yet.

Cite

The paper itself, not this page. BibTeX from the library's record of the work. DOI 10.18653/v1/w19-2703 · arXiv 1808.10290

@inproceedings{shi2019acquiring,
  title = {{Acquiring Annotated Data with Cross-lingual Explicitation for Implicit Discourse Relation Classification}},
  author = {Shi, Wei and Yung, Frances and Demberg, Vera},
  booktitle = {Proceedings of the Workshop on Discourse Relation Parsing and Treebanking 2019},
  year = {2019},
  doi = {10.18653/v1/w19-2703},
  eprint = {1808.10290},
  archivePrefix = {arXiv},
  url = {https://doi.org/10.18653/v1/w19-2703}
}

References

31 references, 19 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. racy, as the country itself, its people and its civil society Shima Gerani, Yashar Mehdad, Giuseppe Carenini, Raymond T Ng, and Bita Nejat. 2014. Abstractive summarization of product reviews using discourse structure. In Proceedings of the 2014 Conference on Empirical Methods in Natural Language Pro… find
  2. Czech back-translation: we all know that Russia cannot return to the path of freedom and democracy there, but Russia itself, its people and civil society. Francisco Guzmán, Shafiq Joty, Lluı́s Màrquez, and Preslav Nakov. 2014. Using discourse structure improves machine translation evaluation. In P… find
  3. We compare the explicitations obtained from translations into three different languages, and find Peter Jansen, Mihai Surdeanu, and Peter Clark. 2014. Discourse complements lexical semantics for non-factoid answer reranking. In Proceedings of the 52nd Annual Meeting of the Association for Computatio… find
  4. Emily Pitler, Annie Louis, and Ani Nenkova. 2009. Automatic sense prediction for implicit discourse relations in text. In Proceedings of the 47th Annual Meeting of the Association for Computational Linguistics, pages 683–691, Suntec, Singapore. Association for Computational Linguistics. find our card
  5. Yangfeng Ji and Jacob Eisenstein. 2015. One vector is not enough: Entity-augmented distributed semantics for discourse relations. Transactions of the Association for Computational Linguistics, 3:329–344. find our card
  6. Emily Pitler, Mridhula Raghupathy, Hena Mehta, Ani Nenkova, Alan Lee, and Aravind K Joshi. 2008. Easily identifiable discourse relations. Technical Reports (CIS), page 884. find our card
  7. Yangfeng Ji, Gholamreza Haffari, and Jacob Eisenstein. 2016. A latent variable recurrent neural network for discourse relation language models. In Proceedings of the 2016 Conference of the North American Chapter of the Association for Computational Linguistics on Human Language Technology, pages 332… find
  8. Rashmi Prasad, Nikhil Dinesh, Alan Lee, Eleni Milt-sakaki, Livio Robaldo, Aravind K. Joshi, and Bonnie L. Webber. 2008. The penn discourse treebank 2.0. In LREC, Marrakech, Morocco. European Language Resources Association. find our card
  9. Lianhui Qin, Zhisong Zhang, and Hai Zhao. 2016. Implicit discourse relation recognition with context-aware character-enhanced embeddings. In Proceedings of COLING 2016, the 26th International Conference on Computational Linguistics. find our card
  10. Philipp Koehn. 2005. Europarl: A parallel corpus for statistical machine translation. In MT summit, volume 5, pages 79–86. find
  11. Lianhui Qin, Zhisong Zhang, Hai Zhao, Zhiting Hu, and Eric P. Xing. 2017. Adversarial connective-exploiting networks for implicit discourse relation classification. In Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 1006– 1017, V… find our card
  12. Philipp Koehn, Hieu Hoang, Alexandra Birch, Chris Callison-Burch, Marcello Federico, Nicola Bertoldi, Brooke Cowan, Wade Shen, Christine Moran, Richard Zens, et al. 2007. Moses: Open source toolkit for statistical machine translation. In Proceedings of the 45th annual meeting of the ACL on interacti… find
  13. Kerstin Kunz and Ekaterina Lapshinova-Koltunski. 2015. Cross-linguistic analysis of discourse variation across registers. Nordic Journal of English Studies, 14(1):258–288. find
  14. Hannah Rohde, Anna Dickinson, Chris Clark, Annie Louis, and Bonnie Webber. 2015. Recovering discourse relations: Varying influence of discourse adverbials. In Proceedings of the First Workshop on Linking Computational Models of Lexical, Sentential and Discourse-level Semantics, pages 22–31. find our card
  15. Ziheng Lin, Min-Yen Kan, and Hwee Tou Ng. 2009. Recognizing implicit discourse relations in the penn discourse treebank. In Proceedings of the 2009 Conference on Empirical Methods in Natural Language Processing, pages 343–351, Singapore. Association for Computational Linguistics. find our card
  16. Attapol Rutherford, Vera Demberg, and Nianwen Xue. 2017. A systematic study of neural discourse models for implicit discourse relation. In Proceedings of the 15th Conference of the European Chapter of the Association for Computational Linguistics: Volume 1, Long Papers, volume 1, pages 281–291. find our card
  17. Ziheng Lin, Hwee Tou Ng, and Min-Yen Kan. 2014. A pdtb-styled end-to-end discourse parser. Natural Language Engineering, 20(02):151–184. find
  18. Attapol Rutherford and Nianwen Xue. 2014. Discovering implicit discourse relations through brown cluster pair representation and coreference patterns. In Proceedings of the 14th Conference of the European Chapter of the Association for Computational Linguistics, pages 645–654. Association for Comput… find our card
  19. Thomas Meyer, Najeh Hajlaoui, and Andrei Popescu-Belis. 2015. Disambiguating discourse connectives for statistical machine translation. IEEE Transactions on Audio, Speech, and Language Processing, 23(7):1184–1197. find
  20. Attapol Rutherford and Nianwen Xue. 2015. Improving the inference of implicit discourse relations via classifying explicit discourse connectives. In Proceedings of the 2015 Conference of the North American Chapter of the Association for Computational Linguistics on Human Language Technology, pages 7… find our card
  21. Joonsuk Park and Claire Cardie. 2012. Improving implicit discourse relation recognition through feature set optimization. In Proceedings of the 13th Annual Meeting of the Special Interest Group on Discourse and Dialogue, pages 108–112. Association for Computational Linguistics. find our card
  22. Wei Shi and Vera Demberg. 2017. On the need of cross validation for discourse relation classification. In Proceedings of the 15th Conference of the European Chapter of the Association for Computational Linguistics, pages 150–156. Association for Computational Linguistics. find our card
  23. Wei Shi and Vera Demberg. 2018. Learning to explicitate connectives with seq2seq network for implicit discourse relation classification. arXiv preprint arXiv:1811.01697. find our card
  24. Wei Shi, Frances Yung, Raphael Rubino, and Vera Demberg. 2017. Using explicit discourse connectives in translation for implicit discourse relation classification. In Proceedings of the Eighth International Joint Conference on Natural Language Processing (Volume 1: Long Papers), volume 1, pages 484–4… find our card
  25. Caroline Sporleder and Alex Lascarides. 2008. Using automatically labelled examples to classify rhetorical relations: An assessment. Natural Language Engineering, 14(3):369–416. find
  26. Suzan Verberne, Lou Boves, Nelleke Oostdijk, and Peter-Arno Coppen. 2007. Evaluating discourse-based answer extraction for why-question answering. In Proceedings of the 30th annual international ACM SIGIR conference on Research and development in information retrieval, pages 735–736. ACM. find
  27. Xun Wang, Sujian Li, Jiwei Li, and Wenjie Li. 2012. Implicit discourse relation recognition by selecting typical training examples. In Proceedings of the 24th International Conference on Computational Linguistics (COLING-2012), pages 2757–2772. find our card
  28. Nianwen Xue, Hwee Tou Ng, Sameer Pradhan, Rashmi Prasad, Christopher Bryant, and Attapol Rutherford. 2015. The conll-2015 shared task on shallow discourse parsing. In Proceedings of the CoNLL-15 Shared Task, pages 1–16. Association for Computational Linguistics. find our card
  29. Nianwen Xue, Hwee Tou Ng, Attapol Rutherford, Bonnie Webber, Chuan Wang, and Hongmin Wang. 2016. find
  30. Conll 2016 shared task on multilingual shallow discourse parsing. In Proceedings of the CoNLL-16 shared task, pages 1–19. Association for Computational Linguistics. find our card
  31. Yasuhisa Yoshida, Jun Suzuki, Tsutomu Hirao, and Masaaki Nagata. 2014. Dependency-based discourse parser for single-document summarization. In Proceedings of the 2014 Conference on Empirical Methods in Natural Language Processing (EMNLP), pages 1834–1839. find our card

↑ back to top