An Integrated Model for Anaphora Resolution

Summary

The paper addresses pronominal anaphora resolution. Mitkov argues that approaches covering a whole language fall short, and that success is likelier when work is restricted to one sublanguage (here, computer science texts) and many knowledge sources are combined. He proposes an architecture where syntactic, semantic, domain, discourse and heuristic modules operate over a shared model of the discourse. The syntactic and semantic modules mostly prune candidates using agreement, c-command, animacy and verb semantics. The domain, heuristic and especially discourse modules suggest antecedents.

The main contribution is a way to find and follow the center of each sentence, including the first sentence of a segment, which earlier centering algorithms took as given. Preferences for the center were derived empirically from computer science writing. They are combined by a Bayesian inference engine, written in Prolog, that revises the probability that a phrase is the center as evidence accumulates. When constraints leave several candidates, the previous sentence's center is picked. The model is shown on worked examples only, and the author calls the results preliminary. Its significance lies in being an early knowledge-rich, domain-specific design for anaphora resolution.

Key points

Center heuristics come from a study of 30 computer science sources totalling over 1000 pages, yielding 11 sublanguage-specific rules. Examples include favouring domain concepts, the object after verbs like "discuss" or "illustrate", repeated NPs, NPs in section headings, and definite NPs.
The center was the subject in about 73% of cases, the object in 25%, and the whole verb phrase in 2%. This gives the a priori priority order subject > object > verb phrase.
Each candidate phrase is scored with Bayes' theorem against two hypotheses (center vs. not center). Observed "symptoms" update the probability one at a time. The engine is a Prolog program that uses a sublanguage knowledge base and a local knowledge base, with manual input at this stage.
Center tracking extends the centering algorithm of Brennan et al. (1987) by also proposing the initial center. A "referential expression filter" removes non-referential pronouns such as "it is necessary".
No quantitative evaluation is given. A common-sense/world-knowledge module is planned but not implemented, and the ordering of the 11 rules is still open.

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 · Coreference resolution

Cited by

6 papers of our library cite it (of the 140673 whose bibliographies we read); 4 of them have a page here, 2 not yet.

Cite

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

@inproceedings{mitkov1994integrated,
  title = {{An Integrated Model for Anaphora Resolution}},
  author = {Mitkov, Ruslan},
  booktitle = {COLING 1994 Volume 2: The 15th International Conference on Computational Linguistics},
  year = {1994},
  doi = {10.3115/991250.991342},
  url = {https://doi.org/10.3115/991250.991342}
}

References

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

Ch. Aonc, D. McKee - Language-independent anaphora resolution system for understanding multilingual texts. ProceeAings of the 31st Annual Meeting of the ACL, The Olfio State University, Colmnbus, Ohio, 1993 find
J. A l l e n Natural language understanding. The Benjamin/Cummings Publishing Company Inc., 1987 find
S. Brennml, M. Fridman, C. Pollard - A centering approach to pronouns. Proceedings of the 25th Annual Meeting of file ACL, Statfford, CA, 1987 find
J. Carbonell, R. 13town - Anaphora resolution: a multi-strategy approach. Proceedings of the 12. International Colfferencc on Computational Linguistics COLING'88, Budapest, 1988 find our card
1). Dahl, C. Ball - Reference resolution in PUNDIT. Research Report CAIT-SLS-9004, March 1990. Center for Advanced Information Teclmology, Paoli, PA 9301 find
R. Frederking, M. Gchrke - Resolving anaphoric references in a DRT-based dialogue system: Part 2: Focus at~l Taxonomic inference. Siemens AG, WlSBER, Bericht Nr. 17, 1987 find
B. Grosz, C. Sidncr - Attention, Intention and the Structure of Discourse. Computalional Linguistics, Vol. 12, 1986 [Hayes 8111 P.J. Hayes - Anaphorafor limited domain systems. Proceedings of the 7th IJCAI, V~mcouver, Canada, 1981 find
G. Hirst - Anaphora in natural language understanding. Berlin Springer Verlag, 1981 find
J. IIobbs - Resolving pronoun refere~es. Lingua, Vol. 44, 1978 [lngria & StaUm'd 8911 R. lngria, D. Stallard - A computational mechanism for pronominal reference. Proceedings of the 27th Annual Meeting of the ACL, Vancouver, British Columbia, 1989 find
R. Mitkov - A knowledge-basedand sublanguage-oriented approach for anaphora resolution. Proceedings of the Pacific Asia Conference on Formal and Computational Linguistics, Taipei, 1993 [Preug et al. 94l PrcuB S., Schmitz B., Hauenschild C., Umbach C. Anaphora Resolution in Machine Translation. In W.… find our card
1~. Rich, S. lalpcrFoy - An architecture for anaphora resolution. Proceedings of the Second Conference on Applitxt Natnral l~anguagc Processing, Austin, Texas, 1988 [Robert 8911 M. Robert - RSsolution de formes pronominales dans l'interface d'interogation d'une base de donndes. Th6sc de doctorat. Fa… find
M. Walker -- Evaluating discourse processing algorithms. Proceedings of the 27th Annual Meeting of the ACL, Vancouver, Colmnbia, 1989 find

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/papers/mitkov-1994-integrated-model-anaphora-resolution/

📄 An Integrated Model for Anaphora Resolution