Elamur Research⌕
☰
home / papers / 2015 / Improving the Inference of Implicit Discourse Relations via Classifying Explicit Discourse Connectives

Improving the Inference of Implicit Discourse Relations via Classifying Explicit Discourse Connectives

Summary

Implicit discourse relations, which lack an overt connective, are the weakest part of PDTB-style discourse parsing because annotated training data is scarce. One remedy is to harvest explicit relations from raw text, remove the connective, and treat the result as extra implicit examples. Earlier attempts at this failed because many connectives change the meaning of a relation when removed. Rutherford and Xue address this by sorting connectives into classes using two statistics computed on the Penn Discourse Treebank 2.0. The first is the omission rate, meaning how often a connective appears as implicit. The second is the context differential, the Jensen-Shannon Divergence between word distributions of a connective's arguments in explicit and implicit uses. They then extract weakly labeled instances from English Gigaword only for "freely omissible" connectives. Adding these instances improves classification of the four top-level PDTB senses over a strong baseline without any new features, while other connective classes do not help. The work shows that distant supervision for implicit relations is viable when the data is filtered carefully.

Key points

  • Data: PDTB 2.0 is used to compute the selection statistics. English Gigaword v3 (about 2.9 billion words of newswire, 1994–2006) supplies the weakly labeled data.
  • Extraction: The authors used the NLTK sentence segmenter, two regular-expression patterns that mirror implicit-relation argument shapes, and filtering rules (for example, excluding "because of"). This yielded more than 200,000 instances.
  • Connective classes: Connectives were grouped as freely omissible, omissible, alternating I and II, and non-omissible.
  • Results with Gigaword data alone: Freely omissible connectives reached a mean accuracy of 0.505, higher than any other class or all classes combined (0.490).
  • Results with implicit PDTB data added: Freely omissible connectives scored 0.571, compared with 0.527–0.554 for the other settings. Gigaword data surpasses the baseline and explicit-PDTB data once roughly 12,000 weakly labeled instances are added. Covering all four senses was not sufficient on its own to improve results; both selection criteria were needed.

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

Cited by

41 papers of our library cite it (of the 140673 whose bibliographies we read); 26 of them have a page here, 15 not yet.

Cite

The paper itself, not this page. BibTeX from the library's record of the work. DOI 10.3115/v1/n15-1081

@inproceedings{rutherford2015improving,
  title = {{Improving the Inference of Implicit Discourse Relations via Classifying Explicit Discourse Connectives}},
  author = {Rutherford, Attapol and Xue, Nianwen},
  booktitle = {Proceedings of the 2015 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies},
  year = {2015},
  doi = {10.3115/v1/n15-1081},
  url = {https://doi.org/10.3115/v1/n15-1081}
}

References

34 references, 13 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Or Biran and Kathleen McKeown. 2013. Aggregated word pair features for implicit discourse relation disambiguation. In Proceedings of the 51st Annual Meeting of the Association for Computational Linguistics, pages 69–73. The Association for Computational Linguistics. find our card
  2. Steven Bird. 2006. Nltk: the natural language toolkit. In Proceedings of the COLING/ACL on Interactive presentation sessions, pages 69–72. Association for Computational Linguistics. find
  3. Vanessa Wei Feng and Graeme Hirst. 2012. Text-level discourse parsing with rich linguistic features. In Proceedings of the 50th Annual Meeting of the Association for Computational Linguistics: Long Papers-Volume 1, pages 60–68. Association for Computational Linguistics. find our card
  4. Bruce Fraser. 1996. Pragmatic markers. Pragmatics, 6:167–190. find
  5. Bruce Fraser. 2006. Towards a theory of discourse markers. Approaches to discourse particles, 1:189–204. find
  6. D Graff, J Kong, K Chen, and K Maeda. 2007. English gigaword third edition, 2007, ldc 2007t07. find
  7. Ben Hutchinson. 2005. Modelling the similarity of discourse connectives. In Proceedings of the the 27th Annual Meeting of the Cognitive Science Society (CogSci2005). find
  8. Lillian Lee. 2001. On the effectiveness of the skew divergence for statistical language analysis. In Artificial Intelligence and Statistics, volume 2001, pages 65–72. find
  9. Junyi Jessy Li and Ani Nenkova. 2014. Reducing sparsity improves the recognition of implicit discourse relations. In Proceedings of the 15th Annual Meeting of the Special Interest Group on Discourse and Dialogue (SIGDIAL), pages 199–207, Philadelphia, PA, U.S.A., June. Association for Computational… find our card
  10. Jiwei Li, Rumeng Li, and Eduard Hovy. 2014. Recursive deep models for discourse parsing. In Proceedings of the 2014 Conference on Empirical Methods in Natural Language Processing (EMNLP), pages 2061–2069. find our card
  11. Ziheng Lin, Hwee Tou Ng, and Min-Yen Kan. 2010. A PDTB-Styled End-to-End Discourse Parser. arXiv.org, November. find
  12. N Ljubesic, Damir Boras, Nikola Bakaric, and Jasmina Njavro. 2008. Comparing measures of semantic similarity. In Information Technology Interfaces, 2008. find
  13. ITI 2008. 30th International Conference on, pages 675– 682. IEEE. find
  14. William C Mann and Sandra A Thompson. 1988. Rhetorical structure theory: Toward a functional theory of text organization. Text, 8(3):243–281. find our card
  15. Micol Marchetti-Bowick and Nathanael Chambers. 2012. Learning for microblogs with distant supervision: Political forecasting with twitter. In Proceedings of the 13th Conference of the European Chapter of the Association for Computational Linguistics, pages 603–612. Association for Computational Ling… find
  16. Daniel Marcu and Abdessamad Echihabi. 2002. An unsupervised approach to recognizing discourse relations. In Proceedings of the 40th Annual Meeting on Association for Computational Linguistics, pages 368–375. Association for Computational Linguistics. find our card
  17. Andrew Kachites McCallum. 2002. Mallet: A machine learning for language toolkit. http://www.cs.umass.edu/ mccallum/mallet. ↗
  18. Mike Mintz, Steven Bills, Rion Snow, and Dan Jurafsky. 2009. Distant supervision for relation extraction without labeled data. In Proceedings of the Joint Conference of the 47th Annual Meeting of the ACL and the 4th International Joint Conference on Natural Language Processing of the AFNLP: Volume 2… find
  19. 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
  20. Gary Patterson and Andrew Kehler. 2013. Predicting the presence of discourse connectives. In Proceedings of the Conference on Empirical Methods in Natural Language Processing. Association for Computational Linguistics. find our card
  21. 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
  22. Emily Pitler, Annie Louis, and Ani Nenkova. 2009. Automatic sense prediction for implicit discourse relations in text. In Proceedings of the Joint Conference of the 47th Annual Meeting of the ACL and the 4th International Joint Conference on Natural Language Processing of the AFNLP: Volume 2-Volume… find our card
  23. Rashmi Prasad, Eleni Miltsakaki, Nikhil Dinesh, Alan Lee, Aravind Joshi, Livio Robaldo, and Bonnie L Webber. 2007. The penn discourse treebank 2.0 annotation manual. find
  24. Rashmi Prasad, Nikhil Dinesh, Alan Lee, Eleni Miltsakaki, Livio Robaldo, Aravind K Joshi, and Bonnie L Webber. 2008. The penn discourse treebank 2.0. In LREC. Citeseer. find our card
  25. Ellen Riloff, Rosie Jones, et al. 1999. Learning dictionaries for information extraction by multi-level bootstrapping. In AAAI/IAAI, pages 474–479. find
  26. Attapol T. 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 (EACL 2014), Gothenburg, Sweden, Ap… find our card
  27. Lawrence Schourup. 1999. Discourse markers. Lingua, 107(3):227–265. find
  28. Michael Speriosu, Nikita Sudan, Sid Upadhyay, and Jason Baldridge. 2011. Twitter polarity classification with label propagation over lexical links and the follower graph. In Proceedings of the First workshop on Unsupervised Learning in NLP, pages 53–63. Association for Computational Linguistics. find
  29. Caroline Sporleder and Alex Lascarides. 2008. Using automatically labelled examples to classify rhetorical relations: An assessment. Natural Language Engineering, 14(03):369–416. find
  30. Attapol Thamrongrattanarit, Colin Pollock, Benjamin Goldenberg, and Jason Fennell. 2013. A distant supervision approach for identifying perspectives in unstructured user-generated text. In Proceedings of the Sixth International Joint Conference on Natural Language Processing, pages 922–926, Nagoya,… find
  31. Xun Wang, Sujian Li, Jiwei Li, and Wenjie Li. 2012. Implicit discourse relation recognition by selecting typical training examples. In Proceedings of COLING 2012, pages 2757–2772, Mumbai, India, December. find our card
  32. The COLING 2012 Organizing Committee. find
  33. Florian Wolf, Edward Gibson, Amy Fisher, and Meredith Knight. 2005. The discourse graphbank: A database of texts annotated with coherence relations. Linguistic Data Consortium. find
  34. Limin Yao, Sebastian Riedel, and Andrew McCallum. 2010. Collective cross-document relation extraction without labelled data. In Proceedings of the 2010 Conference on Empirical Methods in Natural Language Processing, pages 1013–1023. Association for Computational Linguistics. find

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/rutherford-2015-improving-inference-implicit-discourse-relations/