Elamur Research⌕
☰
home / papers / 1997 / Towards resolution of bridging descriptions

Towards resolution of bridging descriptions

Summary

The paper investigates how to automatically resolve bridging definite descriptions: noun phrases with "the" whose antecedent has a different head noun, so the reader must infer a link between them. Their data come from the Penn Treebank: 20 articles published in the Wall Street Journal, with definite descriptions already classified by annotators. They test WordNet as a source of general lexical knowledge. Their method looks for synonymy, direct hyponymy, direct or inherited meronymy, or a shared-hypernym ("sister") relation between a description and candidate noun phrases in the preceding five sentences. This first approach yields few correct resolutions and many false positives. Manual inspection explains much of the failure. The authors sort the bridging cases into six classes and find that only about a fifth involve lexical relations WordNet could plausibly encode, and even those are often missing or arranged unexpectedly in its hierarchy. Choosing the nearest candidate greatly reduces spurious links. The study shows how far a generic lexicon can take bridging resolution and outlines next steps for proper names and event-based anchors.

Key points

  • Corpus: The corpus contains 1040 definite descriptions, of which 204 (20%) are bridging. The rest are mostly same-head anaphora (30%) and discourse-new uses (47%).
  • First WordNet search: Across 5481 anchor/description pairs, the search found 240 relations covering 107 descriptions. Only 54 were correct (spanning 34 descriptions), and 186 were false positives.
  • Six bridging classes: The classes are synonymy/hyponymy/meronymy (38, 19%), names (49, 24%), events (40, 20%), inference (37, 18%), compound nouns (25, 12%), and discourse topic (15, 7%).
  • Lexical-relation cases only: For the 38 synonymy/hyponymy/meronymy cases with manually identified anchors, WordNet recall was just 39% (15/38). Structural quirks contribute to this; for example, house sits under housing rather than building, so the link to walls is lost.
  • Improvements: Keeping only the closest candidate anchor cut false positives from 186 to 77, while correct resolutions fell slightly from 34 to 30 descriptions. Inferring entity types for names from cues such as Mr. or Inc. resolved 53% (26/49) of name-anchored cases. Turning verbs into nominalisations is proposed for event-based anchors.

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

Cited by

11 papers of our library cite it (of the 140673 whose bibliographies we read); 7 of them have a page here, 4 not yet.

Cite

The paper itself, not this page. BibTeX from the library's record of the work. DOI 10.3115/976909.979689

@inproceedings{vieira1997resolution,
  title = {{Towards resolution of bridging descriptions}},
  author = {Vieira, Renata and Teufel, Simone},
  booktitle = {35th Annual Meeting of the Association for Computational Linguistics and 8th Conference of the European Chapter of the Association for Computational Linguistics},
  year = {1997},
  doi = {10.3115/976909.979689},
  url = {https://doi.org/10.3115/976909.979689}
}

References

16 references, 0 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Carter, D. M. 1987. Interpreting Anaphors in .Vat-ural Language Tezts. Ellis Horwood, Chichester. UK. find
  2. Chinchor, N. A. and B. Sundheim. 1995. (MUC) tests of discourse processing. In Proc. AAA[ SS on Empirical Methods in Discourse Interpretation and Generation. pages 21-26, Stanford. find
  3. Clark, H. H. 1977. Bridging. In Johnson-Laird and Wason, eds.. Thinking: Readings in Cognitive Science. Cambridge University Press, Cambridge. find
  4. Clark, H. H. and C. P~. Marshall. 1981. Definite reference and mutual knowledge. In Joshi, Webber and Sag, eds.,Elements of Discourse Understanding. Cambridge University Press, Cambridge. find
  5. Fraurud, K. 1990. Definiteness and the Processing of Noun Phrases in Natural Discourse. Journal of Semantics, 7, pages 39.5-433. find
  6. Grosz, B. J. 1977. The Representation and Use of Focus in Dialogue Understanding. Ph.D. thesis, Stanford University. find
  7. Hawkins, J. A. 1978. Definiteness and Indefiniteness. Croom Helm, London. find
  8. Helm, I. 1982. The Semantics of Definite and Indefinite Noun Phrases. Ph.D. thesis, University of Massachusetts at Amherst. find
  9. Miller, G. et al. 1993. Five papers in WordNet. find
  10. Technical Report CSL Report ~3, Cognitive Science Laboratory, Princeton University. find
  11. Poesio, M. and Vieira. R. 1997. A Corpus based investigation of definite description use. Manuscript, Centre for Cognitive Science, University of Edinburgh. find
  12. Prince, E. 1981. Toward a taxonomy of given/new information. In Cole. ed., Radical Pragmatics. Academic Press. New York, pages '223-255. find
  13. Prince, E. 1992. The ZPG letter: subjects, definete-ness, and information-status. In Thompson and Mann, eds., Discourse description: diverse analyses of a fund raising text. Benjamins. Amsterdam, pages 295-325. find
  14. Sidner, C. L. 1979. Towards a computational theory of definite anaphora comprehension in English discourse. Ph.D. thesis. MIT. find
  15. Strand, K. 1997. A Taxonomy of Linking Relations. Journal of Semantics, forthcoming. find
  16. Vieira, R. and M. Poesio. 1997. Corpus-based processing of definite descriptions. In Botley and McEnery eds., Corpus-based and computational approaches to anaphora. UCL Press. London. find

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/papers/vieira-1997-resolution-bridging-descriptions/