Elamur Research⌕
☰
home / papers / 2010 / Exploiting Scope for Shallow Discourse Parsing

Exploiting Scope for Shallow Discourse Parsing

Summary

The paper addresses shallow discourse parsing on the Penn Discourse Treebank (PDTB). Specifically, it looks at how to find Arg1 of explicit discourse connectives. Arg2 is syntactically tied to the connective and easy to locate, but Arg1 may sit far from it. The authors make two departures from earlier work, which targeted exact spans or argument heads. First, they represent each argument by the sentence that contains it, which also suits extractive summarization. Second, they group connectives by where Arg1 occurs rather than by syntactic class. Intra-sentential connectives keep both arguments in one sentence. Inter-sentential connectives are split into ParaInit (in a paragraph-opening sentence) and ParaNonInit (elsewhere in the paragraph). For ParaNonInit connectives, they propose a filter-rank-evaluate procedure. Scope-based filters shrink the set of preceding sentences in the paragraph that could hold Arg1. Coreference heuristics then pick among any remaining candidates. Applied by hand, the method beats the strong previous-sentence baseline by 3 points absolute. This suggests that discourse scope constraints help with argument identification.

Key points

  • Data: PDTB 2.0 (LDC2008T05) contains 40600 annotated discourse relations, of which 18459 (45%) are explicit connectives. Implicit relations and AltLex are excluded from the study.
  • Arg1 of explicit connectives is located:
    • in the connective's own sentence in 61% of cases,
    • in the immediately preceding sentence in 30%,
    • in a non-adjacent sentence in 9%.
  • Connective taxonomy: intra-sentential versus inter-sentential, with inter-sentential further split by paragraph position into ParaInit and ParaNonInit.
  • Result for ParaNonInit connectives: about a 3% absolute gain over an 83.3% baseline that always chooses the immediately previous sentence as Arg1.
  • Limitations and future work:
    • The algorithm was applied manually.
    • Most errors come from the algorithm itself.
    • The coreference heuristics need refinement.
    • Automation is blocked mainly by the difficulty of detecting direct speech segments automatically, which the scope filters depend on.

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

17 papers of our library cite it (of the 140673 whose bibliographies we read); 5 of them have a page here, 12 not yet.

Cite

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

@inproceedings{prasad2010exploiting,
  title = {{Exploiting Scope for Shallow Discourse Parsing}},
  author = {Prasad, Rashmi and Joshi, Aravind and Webber, Bonnie},
  booktitle = {Proceedings of the Seventh International Conference on Language Resources and Evaluation (LREC'10)},
  year = {2010},
  doi = {10.63317/2zniwupokqvy},
  url = {https://doi.org/10.63317/2zniwupokqvy}
}

References

21 references, 4 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Nicholas Asher and Alex Lascarides. 2003. Logics of conversation. Cambridge University Press. find
  2. Jason Baldridge, Nicholas Asher, and Julie Hunter. 2007. Annotation for and robust parsing of discourse structure on unrestricted texts. Zeitschrift fur Sprachwissenschaft, 26:213–239. find
  3. Dan Cristea, Nancy Ide, Daniel Marcu, and Valentin Tablan. 2000. An empirical investigation of the relation between discourse structure and co-reference. In Proceedings of the 18th conference on Computational linguistics, Morristown, NJ, USA. Association for Computational Linguistics. find
  4. Nikhil Dinesh, Alan Lee, Eleni Miltsakaki, Rashmi Prasad, Aravind Joshi, and Bonnie Webber. 2005. Attribution and the (non)-alignment of syntactic and discourse arguments of connectives. In Proceedings of the ACL Workshop on Frontiers in Corpus Annotation II: Pie in the Sky, Ann Arbor, Michigan. find
  5. Robert Elwell and Jason Baldridge. 2008. Discourse connective argument identification with connective specific rankers. In Proceedings of ICSC-2008. find
  6. Kate Forbes, Eleni Miltsakaki, Rashmi Prasad, Anoop Sarkar, Aravind Joshi, and Bonnie Webber. 2003. D-LTAG system: Discourse parsing with a lexicalized tree adjoining grammar. Journal of Logic, Language and Information, 12(3). find
  7. Barbara J. Grosz, Aravind K. Joshi, and Scott Weinstein. 1995. Centering: A framework for modeling the local coherence of discourse. Computational Linguistics, 21(2):203–225. find our card
  8. Marti A. Hearst. 1997. TextTiling: Segmenting text into multi-paragraph subtopic passages. Computational Linguistics, 23(1):33–64. find
  9. Megumi Kameyama. 1998. Intrasentential centering: A case study. In M.A. Walker, A.K. Joshi, and E.F. Prince, editors, Centering Theory in Discourse, pages 89–112. Oxford University Press, Oxford, U.K. find
  10. Alan Lee, Rashmi Prasad, Aravind Joshi, and Bonnie Webber. 2008. Departures from tree structures in discourse: Shared arguments in the Penn Discourse Treebank. In Proceedings of the Constraints in Discourse III Workshop, Potsdam, Germany. find
  11. Daniel Marcu. 1997. The rhetorical parsing of natural language texts. In Proceedings of the Meeting of the Association for Computational Linguistics. find
  12. Massimo Poesio and Ron Artstein. 2008. Anaphoric annotation in the ARRAU corpus. In Proceedings of 6th International Conference on Language Resources and Evaluation (LREC 2008), Marrakech, Morocco. find our card
  13. Livia Polanyi, Chris Culy, Martin H. van den Berg, Gian Lorenzo Thione, and David Ahn. 2004. A rule based approach to discourse parsing. In Proceedings of the 5th SIGdial Workshop on Discourse and Dialogue, Cambridge, MA. find our card
  14. Rashmi Prasad, Nikhil Dinesh, Alan Lee, Aravind Joshi, and Bonnie Webber. 2007. Attribution and its annotation in the Penn Discourse TreeBank. Traitement Automatique des Langues, Special Issue on Computational Approaches to Document and Discourse, 47(2). find
  15. Rashmi Prasad, Nikhil Dinesh, Alan Lee, Eleni Miltsakaki, Livio Robaldo, Aravind Joshi, and Bonnie Webber. 2008. The Penn Discourse TreeBank 2.0. In Proceedings of 6th International Conference on Language Resources and Evaluation (LREC 2008). find our card
  16. PDTB-Group. 2008. The Penn Discourse TreeBank 2.0 Annotation Manual. Technical Report IRCS-08-01, Institute for Research in Cognitive Science, University of Pennsylvania. find
  17. Violeta Seretan and Dan Cristea. 2002. The use of referential constrains in structuring discourse. In Proceedings of The Third International Conference on Language Resources and Evaluation (LREC 2002), Las Palmas, Spain. find
  18. Bonnie Webber. 2009. Genre distinctions for discourse in the Penn TreeBank. In Proceedings of the Association for Computational Linguistics, Singapore. find
  19. Ralph Weischedel, Sameer Pradhan, Lance Ramshaw, Martha Palmer, Nianwen Xue, Mitchell Marcus, Ann Taylor, Craig Greenberg, Eduard Hovy, Robert Belvin, and Ann Houston. 2007. Ontonotes release 2.0. Technical report, Linguistic Data Consortium. find
  20. Ben Wellner and James Pustejovsky. 2007. Automatically identifiying the arguments of discourse connectives. In Proceedings of EMNLP-CoNLL. find
  21. Ben Wellner. 2009. Sequence Models and Re-ranking Methods for Discourse Parsing. Ph.D. thesis, Brandeis University, Boston, MA. find

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/papers/prasad-2010-exploiting-scope-shallow-discourse-parsing/