Elamur Research⌕
☰
home / papers / 1997 / Resolving bridging references in unrestricted text

Resolving bridging references in unrestricted text

Summary

The paper tackles automatic interpretation of definite descriptions (DDs) in domain-independent written English, focusing on bridging references. These are DDs whose antecedent can only be found through some reasoning, such as a different head noun or a part-whole link. The work builds on a corpus study of Wall Street Journal articles. In that study, one fifth of DDs were bridging cases, which an earlier rule-based prototype did not handle. The authors divide bridging DDs into six processing-oriented classes. Each class gets its own treatment:

  • WordNet lookups for lexical relations
  • heuristics for proper names and compound-noun modifiers
  • a nominalization idea for event anchors

Experiments show WordNet links many DD–anchor pairs but produces far too many false positives. Its coverage of synonymy and meronymy is incomplete, and blind search is computationally expensive. Adding the heuristics to the first prototype raised overall recall from 56% to 65%, with a small precision drop from 84% to 82%. The study matters because it shows what generic lexical resources can and cannot contribute when no hand-built domain knowledge exists.

Key points

  • Corpus: 1040 DDs from parsed WSJ texts (ACL/DCI CD-ROM), split as follows:
    • 204 bridging (20%)
    • 312 same-head anaphoric (30%)
    • 492 larger situation/unfamiliar (47%)
  • Six bridging classes: Syn/Hyp/Mer, Names, Compound Nouns, Events, Discourse Topic and Inference. The last two, plus most event cases, remain beyond the system's reach.
  • WordNet anchor search over a five-sentence window:
    • Some relation was suggested for 107 of 204 bridging DDs, but only 34 had a correct anchor.
    • A Sidner-style stack-based search gave 30 correct and 76 wrong anchors (28% precision).
    • Coordinate-sister links were the least reliable.
  • Manually identified syn/hyp/mer pairs: on 70 such pairs, WordNet reached 46% recall and 84% precision. Synonymy had the weakest recall.
  • Proper names: WordNet supplied entity types for 69% of names, and 53% of DDs referring back to names were resolved.

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

29 papers of our library cite it (of the 140673 whose bibliographies we read); 14 of them have a page here, 15 not yet.

Cite

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

@inproceedings{poesio1997resolving,
  title = {{Resolving bridging references in unrestricted text}},
  author = {Poesio, Massimo and Vieira, Renata and Teufel, Simone},
  booktitle = {Operational Factors in Practical, Robust Anaphora Resolution for Unrestricted Texts},
  year = {1997},
  doi = {10.3115/1598819.1598820},
  url = {https://doi.org/10.3115/1598819.1598820}
}

References

19 references, 1 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Barker, C. 1991. Possessive Descriptions. Ph.D. thesis, University of California at Santa Cruz, Santa Cruz, CA. find
  2. Carter, D. M. 1987. Interpreting Anaphors in Natural Language Texts. Ellis Horwood. find
  3. Clark, H. H. 1977. Bridging. In P. N. Johnson-Laird and P.C. Wason, eds., Thinking: Readings in Cognitive Science. Cambridge University Press. find
  4. Clark, H. H. and C. R. Marshall. 1981. Definite reference and mutual knowledge. In Elements of Discourse Understanding. Cambridge University Press. find
  5. Fraurud, K. 1990. Definiteness and the processing of NPs in natural discourse. Journal of Semantics, 7:395-433. find
  6. Grosz, B. J. and C. L. Sidner. 1986. Attention, intention, and the structure of discourse. Computational Linguistics, 12(3) :175-204. find
  7. Hawkins, J. A. 1978. Definiteness and Indefiniteness. Croom Helm. find
  8. Heim, 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 on WordNet. find
  10. Technical Report CSL Report 43, Cognitive Science Laboratory, Princeton University. find
  11. Poesio, M. 1994. Weak definites. In Proceedings of the Fourth Conference on Semantics and Linguistic Theory, SALT-4. Cornell University Press. find
  12. Poesio, M. and Vieira, R. 1997. A Corpus based investigation of definite description use. Manuscript, Centre for Cognitive Science, University of Edinburgh. find
  13. Prince, E. F. 1981. Toward a taxonomy of given-new information. In P. Cole, ed., Radical Pragmatics. Academic Press, pages 223-256. find
  14. Prince, E. F. 1992. The ZPG letter: subjects, definiteness, and information status. In S. T h o m p - son and W. Mann, eds., Discourse description: diverse analyses of a fund-raising text. John Benjamins, pages 295-325. find
  15. Reinhart, T. 1981. Pragmatics and Linguistics: An Analysis of Sentence Topics Philosophica, 27(1). find
  16. Sidner, C.'L. 1979. Towards a computational theory of definite anaphora comprehension in English discourse. Ph.D. thesis, MIT. find
  17. Strand, K. 1997. A T a x o n o m y of Linking Relations. Journal of Semantics, forthcoming. find
  18. Vieira, R. and M. Poesio. 1997. Processing definite descriptions in corpora. In S. Botley and M. McEnery, eds., Corpus-based and Computational Approaches to Discourse Anaphora. UCL Press. find
  19. Vieira, R. and S. Teufel. 1997. Towards Resolution of Bridging Descriptions. In Proceedings of the find our card

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/poesio-1997-resolving-bridging-references-unrestricted-text/