Elamur Research⌕
☰
home / papers / 2014 / Text-level Discourse Dependency Parsing

Text-level Discourse Dependency Parsing

Paper · 2014 · Sujian Li, Liang Wang, Ziqiang Cao et al.

Summary

The paper addresses text-level discourse parsing, which means organising a whole document into one discourse structure. Earlier systems built RST-style constituency trees, and the authors identify three weaknesses in that approach: production rules for inner spans are hard to define, units at different levels need different features, and approximate search can get stuck in local optima. Their alternative links elementary discourse units (EDUs) directly through head–dependent arcs, with an artificial root unit e0, and drops intermediate spans entirely.

To get training data, they convert RST-DT into a discourse dependency treebank. Nuclei serve as heads, and n-ary relations are binarised. Parsing is graph-based and arc-factored. The Eisner algorithm produces projective trees and the MST algorithm produces non-projective ones. Arc weights are learned with MIRA, a large-margin method, using six feature types.

On RST-DT the parser is competitive with or better than HILDA and other systems, especially for nuclearity and relation labeling. The authors present this as the first use of dependency representations and parsers for discourse. Its main appeal is a simpler framework with lower computational cost.

Key points

  • Resource: a discourse dependency treebank converted from RST-DT. RST-DT uses 110 fine-grained relations grouped into 18 classes. Its 381 n-ary relations were binarised, with the first nucleus taken as head.
  • Methods: Eisner (projective) and MST (non-projective) decoding over arc-factored scores, with weights trained by MIRA. A perceptron variant (Percep-coarse) serves as a comparison.
  • Full-parser results (span / nuclearity / relation):
    • Our-coarse: 82.9 / 73.0 / 60.6
    • Our-fine: 83.4 / 73.8 / 57.8
    • HILDA-manual: 83.0 / 68.4 / 55.3
    • Human agreement: 88.1 / 77.5 / 66.0
  • Relation labeling, Our-coarse vs. Feng & Hirst (2012):
    • MAFS: 0.454 vs. 0.440
    • WAFS: 0.643 vs. 0.607
    • Accuracy: 66.84% vs. 65.30%
    • Majority baseline (always Elaboration): 35.82%
  • The converted treebank contains no non-projective trees, although the formal definition allows them. The authors name non-projective discourse parsing as future work.

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

 Institutions

🎓Peking UniversityBeijing · China100 works · 35,618 citations🎓Hong Kong Polytechnic UniversityHong Kong · China34 works · 4,476 citations

📚 Shelves 4The field of science this work belongs to (OpenAlex fields).Computer Science · 9,086 worksArtificial Intelligence · 8,889 worksArts and Humanities · 9,296 worksLanguage and Linguistics · 9,206 works

🧭 Topics 2Research subjects and languages the work is about.Discourse analysis · 394 worksDependency parsing · 120 works

Cite

Cite as Sujian Li, Liang Wang, Ziqiang Cao and Wenjie Li. 2014. Text-level Discourse Dependency Parsing. Proceedings of the 52nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers).

The paper itself, not this page. BibTeX from the library's record of the work. DOI 10.3115/v1/p14-1003

@inproceedings{li2014text,
  title = {{Text-level Discourse Dependency Parsing}},
  author = {Sujian Li, Liang Wang, Ziqiang Cao and Wenjie Li},
  booktitle = {Proceedings of the 52nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)},
  year = {2014},
  doi = {10.3115/v1/p14-1003},
  url = {https://doi.org/10.3115/v1/p14-1003}
}

References

28 references, 14 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Jason Baldridge and Alex Lascarides. 2005. Probabilistic Head-driven Parsing for Discourse Structure. In Proceedings of the Ninth Conference on Computational Natural Language Learning, pages 96– 103. find
  2. Steven Bird, Ewan Klein, and Edward Loper. 2009. Natural Language Processing with Python — Analyzing Text with the Natural Language Toolkit. O’Reilly. find
  3. Lynn Carlson, Daniel Marcu, and Mary E. Okurowski. 2001. Building a Discourse-tagged Corpus in the Framework of Rhetorical Structure Theory. Proceedings of the Second SIGdial Workshop on Discourse and Dialogue-Volume 16, pages 1–10.
  4. Yoeng-Jin Chu and Tseng-Hong Liu. 1965. On the Shortest Arborescence of a Directed Graph, Science Sinica, v.14, pp.1396-1400. find
  5. Koby Crammer and Yoram Singer. 2003. Ultraconservative Online Algorithms for Multiclass Problems. JMLR. find
  6. Jack Edmonds. 1967. Optimum Branchings, J. Research of the National Bureau of Standards, 71B, pp.233-240. find
  7. Jason Eisner. 1996. Three New Probabilistic Models for Dependency Parsing: An Exploration. In Proc. COLING.
  8. Vanessa Wei Feng and Graeme Hirst. Text-level Discourse Parsing with Rich Linguistic Features, Proceedings of the 50th Annual Meeting of the Association for Computational Linguistics, pages 60–68, Jeju, Republic of Korea, 8-14 July 2012.
  9. Hugo Hernault, Danushka Bollegala, and Mitsuru Ishizuka. 2010a. A Semi-supervised Approach to Improve Classification of Infrequent Discourse Relations Using Feature Vector Extension. In Proceedings of the 2010 Conference on Empirical Methods in Natural Language Processing, pages 399–409, Cambridge,…
  10. Hugo Hernault, Helmut Prendinger, David A. duVerle, and Mitsuru Ishizuka. 2010b. HILDA: A Discourse Parser Using Support Vector Machine Classification. Dialogue and Discourse, 1(3):1–33.
  11. Shafiq Joty, Giuseppe Carenini and Raymond T. Ng. A Novel Discriminative Framework for Sentence-level Discourse Analysis. EMNLP-CoNLL '12 Proceedings of the 2012 Joint Conference on Empirical Methods in Natural Language Processing and Computational Natural Language Learning Stroudsburg, PA, USA. find
  12. Huong LeThanh, Geetha Abeysinghe, and Christian Huyck. 2004. Generating Discourse Structures for Written Texts. In Proceedings of the 20th International Conference on Computational Linguistics, pages 329– 335. find
  13. 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 Method in Natural Language Processing, Vol. 1, EMNLP’09, pages 343-351.
  14. William Mann and Sandra Thompson. 1988. Rhetorical Structure Theory: Toward a Functional Theory of Text Organization. Text, 8(3):243–281.
  15. Daniel Marcu. 2000. The Theory and Practice of Discourse Parsing and Summarization. MIT Press, Cambridge, MA, USA. find
  16. Ryan McDonald, Koby Crammer, and Fernando Pereira. 2005a. Online Large-Margin Training of Dependency Parsers, 43rd Annual Meeting of the Association for Computational Linguistics (ACL 2005) .
  17. Ryan McDonald, Fernando Pereira, Kiril Ribarov, and Jan Hajic. 2005b. Non-projective Dependency Parsing using Spanning Tree Algorithms, Proceedings of HLT/EMNLP 2005.
  18. Emily Pitler, Annie Louis, and Ani Nenkova. 2009. Automatic Sense Prediction for Implicit Discourse Relations in Text, In Proc. of the 47th ACL. pages 683-691.
  19. Rashmi Prasad, Eleni Miltsakaki, Nikhil Dinesh, Alan Lee, Aravind Joshi, Livio Robaldo, and Bonnie Webber. 2007. The Penn Discourse Treebank 2.0 Annotation Manual. The PDTB Research Group, December. find
  20. David Reitter. 2003. Simple Signals for Complex Rhetorics: On Rhetorical Analysis with Rich-feature Support Vector Models. LDV Forum, 18(1/2):38–52. find
  21. Kenji Sagae. 2009. Analysis of discourse structure with syntactic dependencies and data-driven shift-reduce parsing. In Proceedings of the 11th International Conference on Parsing Technologies, pages 81-84. find
  22. Radu Soricut and Daniel Marcu. 2003. Sentence level discourse parsing using syntactic and lexical information. In Proceedings of the 2003 Conference of the North American Chapter of the Association for Computational Linguistics on Human Language Technology, Volume 1, pages 149–156.
  23. Rajen Subba and Barbara Di Eugenio. 2009. An effective discourse parser that uses rich linguistic information. In Proceedings of Human Language Technologies: The 2009 Annual Conference of the North American Chapter of the Association for Computational Linguistics, pages 566–574.
  24. Robert Endre Tarjan, 1977. Finding Optimum Branchings, Networks, v.7, pp.25-35. find
  25. Ben Taskar, Carlos Guestrin and Daphne Koller. 2003. Max-margin Markov Networks. In Proc. NIPS. find
  26. Bonnie Webber. 2004. D-LTAG: Extending Lexicalized TAG to Discourse. Cognitive Science, 28(5):751–779. find
  27. Wen Ting Wang, Jian Su and Chew Lim Tan. 2010. Kernel based Discourse Relation Recognition with Temporal Ordering Information, In Proc. of ACL’10. pages 710-719.
  28. Ben Wellner, James Pustejovsky, Catherine Havasi, Anna Rumshisky and Roser Sauri. 2006. Classification of Discourse Coherence Relations: an Exploratory Study Using Multiple Knowledge Sources. In Proc.of the 7th SIGDIAL Workshop on Discourse and Dialogue. pages 117-125.

Updated

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/li-2014-text-level-discourse-dependency-parsing/