Elamur Research⌕
☰
home / papers / 2004 / A Rule Based Approach to Discourse Parsing

A Rule Based Approach to Discourse Parsing

Summary

The paper tackles three obstacles that kept the Linguistic Discourse Model (LDM), a semantics-oriented theory of discourse structure, from being implemented: splitting text into basic units, choosing where a new unit attaches in the growing discourse tree, and choosing the relation it attaches with. The authors propose the Unified LDM (U-LDM). It grounds segmentation in semantics: a segment carries at most one event or state, or performs a single discourse function. The syntactic constructions that express this are then identified, covering clauses, participial and infinitival modifiers, appositives, parentheticals, conjunctions and cue words. Attachment is handled by symbolic syntactic, semantic and lexical rules. These rules rank candidate nodes on the right edge of the tree and then link the incoming unit with a coordination, subordination or n-ary relation. When several sites remain possible, ambiguous parses are kept. The approach is demonstrated in PALSUMM, a symbolic summarizer that prunes discourse trees. No large-scale quantitative evaluation is reported. The work matters because it shows that discourse parsing can be rule-based and interpretable, and it supplies structures that could later train statistical systems.

Key points

  • The LDM represents discourse as an open right tree built from three node types: coordination (C), subordination (S) and n-ary (N). Only nodes on the right edge stay open for continuation and anaphora resolution.
  • U-LDM segments are split into content segments (eventualities, interpolations, fragments) and operator segments (conjunctions, scene-setting preposed modifiers, cue words). Segments can be discontinuous, as with appositives or non-restrictive relatives.
  • PALSUMM uses Xerox Linguistic Environment (XLE) f-structures for sentence parsing. It draws on WordNet plus a custom domain ontology to build BDU-trees and attach them via discourse rules. Summaries come from cutting the tree at a chosen embedding depth or from salience-guided partial trees.
  • The system was applied to a corpus of 300 FX Palo Alto Laboratory technical reports. A small comparative study rated its summaries as clearly more readable than MEAD output.
  • The authors contrast LDM trees with RST trees. LDM trees are constituent trees with essentially two node kinds, while RST trees are dependency structures with many link types, which makes tree-based summarization simpler under LDM.

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

28 papers of our library cite it (of the 140673 whose bibliographies we read); 5 of them have a page here, 23 not yet.

Cite

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

@inproceedings{polanyi2004rule,
  title = {{A Rule Based Approach to Discourse Parsing}},
  author = {Polanyi, Livia and Culy, Chris and Berg, Martin van den and Thione, Gian Lorenzo and Ahn, David},
  booktitle = {Proceedings of the 5th SIGdial Workshop on Discourse and Dialogue at HLT-NAACL 2004},
  year = {2004},
  url = {https://aclanthology.org/W04-2322/}
}

References

17 references, 2 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Nicholas Asher. 1993. Reference to Abstract Objects in English: A Philosophical Semantics for Natural Language Metaphysics. Kluwer Academic Publishers. find
  2. Lynn Carlson, Daniel Marcu, and Mary Ellen Okurowski. To appear. Building a Discourse-Tagged Corpus in the Framework of Rhetorical Structure Theory. In Current Directions in Discourse and Dialogue, Jan van Kuppevelt and Ronnie Smith eds. Kluwer Academic Publishers. find our card
  3. Katherine M. 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 Language, Logic and Information, 12(3). find
  4. Barbara Grosz and Candace Sidner. 1986. Attention, Intention and the Structure of Discourse. Computational Linguistics 12:175-204.. find
  5. Hans Kamp. 1981. A theory of truth and semantic representation.” In Formal Methods in the Study of Language. Jeroen A. G. Groenendijk, Theo Janssen, and Martin Stokhof (eds.). Amsterdam: Mathematisch Centrum, 277-322. find
  6. Hans Kamp and Uwe Reyle. 1993. From Discourse to Logic: Introduction to Model-theoretic Semantics of Natural Language, Formal Logic and Discourse Representation Theory. Kluwer Academic Publishers, Dordrecht, The Netherlands. find
  7. William C. Mann and Sandra A. Thompson. 1988. Rhetorical Structure Theory: Towards a Functional Theory of Text Organization. Text 8(3)243-281. find
  8. Daniel Marcu. 1999. Discourse trees are good indicators of importance in text. In Advances in Automatic Text Summarization. I. Mani and Mark Maybury (Eds.), 123-136, The MIT Press. find
  9. Daniel Marcu. 2000. The Theory and Practice of Discourse Parsing and Summarization. The MIT Press. Cambridge, MA. find
  10. Daniel Marcu. 2003. Automatic Abstracting, Encyclopedia of Library and Information Science, 245-256, 2003. find
  11. Livia Polanyi. 1988. A Formal Model of Discourse Structure. Journal of Pragmatics 12: 601-639. find
  12. Livia Polanyi and Martin van den Berg. 1996. Discourse Structure and Discourse Interpretation. In Proceedings of the 10 th Amsterdam Colloquium on Formal Semantics. University of Amsterdam. find
  13. Livia Polanyi and Remko Scha. 1984. A syntactic approach to discourse semantics. In Proceedings of COLING 6. Stanford, CA. 413-419. find
  14. Dragomir Radev, Timothy Allison, Sasha Blair-Goldensohn, and John Blitzer, Arda Çelebi, Elliott Drabek, Wai Lam, Danyu Liu, Hong Qi, Horacio Saggion, Simone Teufel, Michael Topper, Adam Winkel. 2003. “The MEAD Multidocument Summarizer”. http://www.summarization.com/mead/ ↗
  15. Radu Soricut and Daniel Marcu. 2003. Sentence Level Discourse Parsing using Syntactic and Lexical Information. In Proceedings of HLT/NAACL’03, May 27-June 1, Edmonton, Canada Bonnie Webber.1991. Structure and Ostension in the Interpretation of discourse Deixis. In Language and Cognitive Processes, 6… find our card
  16. Bonnie Webber and Aravind Joshi. 1998. Anchoring a Lexicalized Tree-Adjoining Grammar for Discourse. ACL/COLING Workshop on Discourse Relations and Discourse Markers, Montreal, Canada. find
  17. Alec Wilkinson. 2003. Talk of the Town, September 26, 2003, New Yorker find

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/polanyi-2004-rule-based-approach-discourse-parsing/