Elamur Research⌕
☰
home / papers / 2003 / Automatic Pronominal Anaphora Resolution in English Texts

Automatic Pronominal Anaphora Resolution in English Texts

Summary

The paper addresses automatic resolution of English third-person pronouns (nominative, reflexive, possessive and objective forms). This step matters for machine translation, summarization and question answering, where unresolved references break coherence. The authors build a fully automatic pipeline that does not need full syntactic parsing. Text is tagged with the TOSCA-ICLE tagger, and base noun phrases are extracted by a finite state machine. Non-referential "it" is filtered out with seven patterns built around modal adjectives and cognitive verbs. Candidate antecedents within a fixed window, in the same or preceding sentences, are pruned by number, gender and animacy agreement. Gender comes from a first-name list and WordNet keywords. Animacy is inferred from WordNet hypernym chains that lead to the person or animal top-level categories. The remaining candidates are ranked by a score that combines preference and constraint heuristics. On random Brown corpus texts the system resolves 77% of anaphors, well above simple recency baselines. The result shows that lightweight lexical resources can support competitive pronoun resolution without a deep parser.

Key points

  • Test data: random multi-genre Brown corpus texts with 14,124 words, 2,970 noun phrases and 530 anaphors. The metric is success rate, defined as correctly resolved anaphors divided by all anaphors.
  • Overall success rate is 77%. By genre it ranges from 69% (Lore) to 80% (Reportage, Editorial). The baselines score 41% (most recent subject) and 17% (most recent agreeing NP).
  • Pleonastic "it" made up about 9% of the pronouns, and the system identified it with 89% accuracy.
  • Resources: WordNet, used for animacy via the {person, human being} and {animal, fauna} unique beginners and for gender keywords. A behindthename.com list supplies 5,661 male and 5,087 female first names.
  • For context, the paper cites earlier figures from [Mitkov, 01] on a different corpus: Mitkov's own approach reached 56.9%, CogNIAC scored 49.72% and the Kennedy & Boguraev algorithm achieved 61.6%. These numbers are not directly comparable with the 77% result.

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: Coreference resolution

Cited by

6 papers of our library cite it (of the 140673 whose bibliographies we read); 1 of them have a page here, 5 not yet.

Cite

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

@inproceedings{liang2003automatic,
  title = {{Automatic Pronominal Anaphora Resolution in English Texts}},
  author = {Liang, Tyne and Wu, Dian-Song},
  booktitle = {Proceedings of Research on Computational Linguistics Conference XV},
  year = {2003},
  doi = {10.30019/ijclclp.200402.0002},
  url = {https://doi.org/10.30019/ijclclp.200402.0002}
}

References

17 references, 8 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Aarts Jan, Henk Barkema and Nelleke Oostdijk (1997), “The TOSCA-ICLE Tagset: Tagging Manual”, TOSCA Research Group for Corpus Linguistics. find
  2. Baldwin, Breck (1997), “CogNIAC: high precision coreference with limited knowledge and linguistic resources”, Proceedings of the ACL'97/EACL'97 workshop on Operational factors in practical, robust anaphora resolution, pp. 38-45. find our card
  3. Bontcheva, Kalina, Marin Dimitrov, Diana Maynard and Valentin Tablan (2002), “Shallow Methods for Named Entity Coreference Resolution”, Proceedings of TRAITEMENT AUTOMATIQUE DES LANGUES NATURELLES (TALN), pp. 24-32. find
  4. Cardie, Claire and Kiri Wagstaff (1999), “Noun Phrase Coreference as Clustering”, Proceedings of the Joint Conference on Empirical Methods in Natural Language Processing and Very Large Corpora. find our card
  5. Chen, Kuang-hua and Hsin-Hsi Chen (1994), “Extracting Noun Phrases from Large-Scale Texts: A Hybrid Approach and Its Automatic Evaluation”, Proceedings of the 32nd ACL Annual Meeting, 1994, pp. 234-241. find
  6. Dagan, Ido and Alon Itai (1990), “Automatic processing of large corpora for the resolution of anaphora references”, Proceedings of the 13th International Conference on Computational Linguistics (COLING'90), Vol. III, 1-3, Helsinki, Finland. find our card
  7. Denber, Michel (1998), “Automatic resolution of anaphora in English”, Technical report, Eastman Kodak Co. find
  8. Evans, Richard and Constantin Orasan (2000), “Improving anaphora resolution by identifying animate entities in texts”, In Proceedings of DAARC-2000. find
  9. Ge, Niyu, John Hale and Eugene Charniak (1998), “A Statistical Approach to Anaphora Resolution”, Proceedings of the Sixth Workshop on Very Large Corpora (COLING-ACL98), pp.161-170. find our card
  10. Kennedy, Christopher and Branimir Boguraev (1996), “Anaphora for everyone: Pronominal anaphora resolution without a parser”, Proceedings of the 16 th International Conference on Computational Linguistics, pp.113-118. find our card
  11. Lappin, Shalom and Herbert Leass (1994), “An Algorithm for Pronominal Anaphora Resolution”, Computational Linguistics, Volume 20, Part 4, pp. 535-561. ↗ our card
  12. Miller, George (1993), “Nouns in WordNet: A Lexical Inheritance System”, Journal of Lexicography, pp. 245-264. find
  13. Mitkov, Ruslan (1998), “Robust pronoun resolution with limited knowledge”, Proceedings of the 18th International Conference on Computational Linguistics (COLING'98)/ACL'98 Conference Montreal, Canada. pp. 869-875. find our card
  14. Mitkov, Ruslan (1999), “Anaphora Resolution: The State of the Art”, Working paper (Based on the COLING'98/ACL'98 tutorial on anaphora resolution) find
  15. Mitkov, Ruslan and Catalina Barbu (2001), “Evaluation tool for rule-based anaphora resolution methods”, Proceedings of ACL'01, Toulouse, 2001. find our card
  16. Mitkov, Ruslan, Richard Evans and Constantin Orasan (2002), “A new, fully automatic version of Mitkov's knowledge-poor pronoun resolution method”, In Proceedings of CICLing- 2000, Mexico City, Mexico. find
  17. Wang, Ning, Chunfa Yuan, K.F. Wang and Wenjie Li (2002), “Anaphora Resolution in Chinese Financial News for Information Extraction”, Proceedings of 4th World Congress on Intelligent Control and Automation, June 2002, Shanghai, pp.2422-2426. find

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/papers/liang-2003-automatic-pronominal-anaphora-resolution-english/