Elamur Research⌕
☰
home / papers / 2011 / Shallow Discourse Parsing with Conditional Random Fields

Shallow Discourse Parsing with Conditional Random Fields

Summary

This paper addresses a gap in discourse parsing. Earlier systems either recovered only argument heads or covered only some connectives. Here the task is to find the complete text spans of both arguments (Arg1 and Arg2) of explicit discourse connectives in the Penn Discourse Treebank (PDTB). The authors assume the connectives are already known and make no prior assumptions about span shape. They frame the problem as token-level sequence labeling, solved by a cascade of conditional random fields (CRFs). Arg2, which is syntactically tied to the connective, is labeled first, and that output then feeds the Arg1 model. The features are lexical, syntactic and semantic, drawn from gold PDTB annotation and Penn Treebank parse trees, including the connective's top-level sense class. The best feature combinations include both syntactic and semantic information. Arg1 within the connective's own sentence is found about as reliably as Arg2. Arg1 in an earlier sentence remains the hardest case. The approach yields a flat, non-hierarchical discourse representation that can support applications such as opinion mining, where exact argument spans matter.

Key points

  • Data: PDTB, about one million words of Wall Street Journal text with 18,459 explicit connective tokens of 100 types. Arg1 sits in the same sentence as the connective in 60.9% of cases. The test split contains 1028 intra-sentential and 617 inter-sentential relations.
  • Arg1 within a sentence (exact match): F1 is about 0.66–0.68, rising to roughly 0.85 under the more lenient overlap metric.
  • Arg1 across sentences: this case depends on a "Prev" feature. With it, exact-match F1 rises from 0.02 to 0.36.
  • Results by connective type: Arg2 exact-match F1 is fairly uniform at 0.78–0.82. Adverbial connectives are the weakest case for Arg1, reaching 0.44 F1 even with Prev.
  • Efficiency and error patterns: pruning the search space by 90% kept training time manageable while dropping only about 5% of Arg1 instances. Recall was consistently below precision. Most errors involved locating exact span boundaries rather than finding the right sentence.

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

27 papers of our library cite it (of the 140673 whose bibliographies we read); 11 of them have a page here, 16 not yet.

Cite

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

@inproceedings{ghosh2011shallow,
  title = {{Shallow Discourse Parsing with Conditional Random Fields}},
  author = {Ghosh, Sucheta and Johansson, Richard and Riccardi, Giuseppe and Tonelli, Sara},
  booktitle = {Proceedings of 5th International Joint Conference on Natural Language Processing},
  year = {2011},
  url = {https://aclanthology.org/I11-1120/}
}

References

23 references, 11 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Rich Caruana and Dayne Freitag. 1994. Greedy attribute selection. In Proceedings of the Eleventh International Conference on Machine Learning. find
  2. 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 Workshop on Frontiers in Corpus Annotations II: Pie in the Sky, pages 29–36, Ann Arbor, Mic… find
  3. Robert Elwell and Jason Baldridge. 2008. Discourse connective argument identification with connective specific rankers. In Proceedings of ICSC-2008, Santa Clara, United States. find
  4. Sucheta Ghosh, Sara Tonelli, Giuseppe Riccardi, and Richard Johansson. 2011. End-to-end discourse parser evaluation. In Proceedings of 5th IEEE International Conference on Semantic Computing, Palo Alto, CA, USA. find
  5. J. S. Urban Hjorth. 1993. Computer Intensive Statistical Methods. Chapman and Hall, London. find
  6. Richard Johansson and Alessandro Moschitti. 2010. Syntactic and semantic structure for opinion expression detection. In Proceedings of the Fourteenth Conference on Computational Natural Language Learning, pages 67–76. find
  7. John Lafferty, Andrew McCallum, and Fernando Pereira. 2001. Conditional random fields: Probabilistic models for segmenting and labeling sequence data. In 18th International Conf. on Machine Learning. Morgan Kaufmann. find
  8. 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. find our card
  9. William Mann and Sara Thompson. 1988. Rhetorical structure theory: Toward a functional theory of text organization. Text, 8(3):243–281. find our card
  10. Mitchell P. Marcus, Beatrice Santorini, and Mary Ann Marcinkiewicz. 1993. Building a Large Annotated Corpus of English: the Penn Treebank. Computational Linguistics, 19(2):313–330. find our card
  11. Guido Minnen, John Carroll, and Darren Pearce. 2001. Applied morphological processing of English. Natural Language Engineering. find
  12. Emily Pitler and Ani Nenkova. 2009. Using syntax to disambiguate explicit discourse connectives in text. In Proceedings of the 47 th Annual Meeting of the Association for Computational Linguistics and the 4 th International Joint Conference on Natural Language Processing. find our card
  13. Emily Pitler, Mridhula Raghupathy, Hena Mehta, Ani Nenkova, Alan Lee, and Aravind Joshi. 2008. Easily identifiable discourse relations. In Proceedings of the 22nd International Conference on Computational Linguistics (Coling 2008), pages 87–90, Manchester, United Kingdom. find our card
  14. 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, pages 683–691, S… find our card
  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 the 6 th International Conference on Languages Resources and Evaluations (LREC 2008), Marrakech, Morocco. find our card
  16. Rashmi Prasad, Aravind Joshi, and Bonnie Webber. 2010. Exploiting scope for shallow discourse parsing. In Proceedings of the Seventh conference on International Language Resources and Evaluation (LREC’10), Valletta, Malta. find our card
  17. Fei Sha and Fernando Pereira. 2003. Shallow parsing with conditional random fields. In Proceedings of HLT/NAACL, pages 213–220. find
  18. Swapna Somasundaran, Galileo Namata, Janyce Wiebe, and Lise Getoor. 2009. Supervised and unsupervised methods in employing discourse relations for improving opinion polarity classification. In Proceedings of the 2009 Conference on Empirical Methods in Natural Language Processing, pages 170–179, Sing… find our card
  19. Sara Tonelli, Giuseppe Riccardi, Rashmi Prasad, and Aravind Joshi. 2010. Annotation of discourse relations for conversational spoken dialogs. In Proceedings of the Seventh conference on International Language Resources and Evaluation (LREC’10), Valletta, Malta. find our card
  20. Bonnie Webber, Markus Egg, and Valia Kordoni. 2010. Discourse Structure and Language Technology. Natural Language Engineering, 1(1):1–49. find
  21. Ben Wellner and James Pustejovsky. 2007. Automatically identifying the arguments of discourse connectives. In Proceedings of the 2007 Joint Conference on Empirical Methods in Natural Language Processing and Computational Natural Language Learning (EMNLP-CoNLL), pages 92–101, Prague, Czech Republic. find
  22. Ben Wellner. 2009. Sequence Models and Ranking Methods for Discourse Parsing. Ph.D. thesis, Brandeis University. find
  23. Hiroyasu Yamada and Yuji Matsumoto. 2003. Statistical dependency analysis with support vector machines. In Proceedings of 8th International Workshop on Parsing Technologies. find our card

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/ghosh-2011-shallow-discourse-parsing-conditional-random/