Elamur Research
☰
home / papers / 1994 / An Algorithm for Pronominal Anaphora Resolution

An Algorithm for Pronominal Anaphora Resolution

Summary

The paper presents RAP (Resolution of Anaphora Procedure), a classic rule-based algorithm for resolving third-person pronouns (including reflexives and reciprocals) using syntactic information from a parser. Candidate antecedents are assigned a salience score built from weighted factors such as recency, grammatical role (subject, existential, direct object, indirect object/oblique), head-noun status and not being inside an adverbial phrase. Salience is halved for each sentence boundary crossed, and role parallelism and cataphora adjust the score. Syntactic filters (agreement and binding-style constraints on co-arguments) remove impossible candidates before the highest-scoring one is chosen. The authors evaluate RAP on technical manuals and compare it with Hobbs' syntactic search algorithm, reporting that the salience approach performs at least as well. The work became a standard reference point for knowledge-poor, transparent pronoun resolution.

How Mova uses it

Effectiveness in Mova

Not tried yet (the coref README marks it explicitly as "not tried"). Planned idea only; not measured.

👨‍🔬💥

Library summary

The Research library's own summary of the paper, in our words — not the paper's text.

Written for the Research library (claude-opus-5-5) from the paper text, in our own words: the license "CC BY-NC-SA 3.0 (ACL Anthology, before 2016)" does not allow redistributing the text itself. Original — aclanthology.org/J94-4002.pdf

Summary

Lappin and Leass tackle the task of linking third person pronouns, reflexives and reciprocals to their noun phrase antecedents, both within a sentence and across sentences. Their system, RAP (Resolution of Anaphora Procedure), is written in Prolog and runs on parses from McCord's Slot Grammar. It uses no semantic reasoning or world knowledge. Instead, it removes impossible candidates with syntactic and agreement filters, flags pleonastic "it", and binds lexical anaphors with a separate algorithm. The remaining candidates are ranked by salience scores. These scores combine grammatical role, role parallelism, frequency of mention, proximity and sentence recency, and they are updated as the discourse unfolds. In a blind test on computer manual text with 360 pronouns, RAP chose the correct antecedent in 86% of cases. This is 4 points better than a Slot Grammar reimplementation of Hobbs' algorithm. Adding Dagan's statistical lexical preferences improved accuracy by only about 2%. The study suggests that inexpensive structural and attentional cues do most of the work. Semantic knowledge can then serve as a secondary step for close calls.

Key points

  • RAP operates on Slot Grammar clausal representations (ESG; a GSG version also exists). It keeps the four preceding sentences in memory and is integrated into the LMT English–German machine translation system to choose the gender of target-language pronouns.
  • Its components are a six-condition syntactic non-coreference filter, morphological agreement checks, pattern-based detection of pleonastic "it" (modal adjectives, cognitive verbs), a binding algorithm for reflexives and reciprocals, and salience weighting computed over equivalence classes of coreferent NPs.
  • The grammatical role hierarchy ranks subjects above other NPs, direct objects above other complements, arguments above adjuncts, and head nouns above their complements.
  • Blind test: 360 pronoun occurrences, 86% accuracy, compared with a 4-point lower score for Hobbs' search procedure on the same sentences.
  • The statistically enhanced variant (RAPSTAT, using Dagan 1992 lexical preferences) gained only about 2%. It mainly helped when the top salience scores were close and a less salient candidate had strong lexical support.

Topics: Coreference resolution

Cite

The paper itself, not this card. BibTeX from the library's record of the work. DOI 10.5555/203987.203989

@misc{lappin1994algorithm,
  title = {{An Algorithm for Pronominal Anaphora Resolution}},
  author = {Lappin, Shalom and Leass, Herbert J.},
  year = {1994},
  doi = {10.5555/203987.203989},
  url = {https://doi.org/10.5555/203987.203989}
}

References

31 references, 3 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Alshawi, Hiyan (1987). Memory and Context for Language Interpretation. Cambridge: Cambridge University Press. find
  2. Asher, Nicholas, and Wada, Hajime (1988). "A computational account of syntactic, semantic and discourse principles for anaphora resolution." Journal of Semantics 6:309-344. find
  3. Bosch, Peter (1988). "Some good reasons for shallow pronoun processing." In Proceedings, IBM Conference on Natural Language Processing. New York: Thornwood. find
  4. Brennan, Susan; Friedman, Marilyn; and Pollard, Carl (1987). "A centering approach to pronouns." In Proceedings, 25th Annual Meeting of the Association for Computational Linguistics, 155-162. find
  5. Carbonell, Jaime, and Brown, Ralf (1988). "Anaphora resolution: A multi-strategy approach." In Proceedings, 12th International Conference on Computational Linguistics, 96-101. find
  6. Dagan, Ido (1992). "Multilingual statistical approaches for natural language disambiguation" (in Hebrew). Doctoral dissertation, Israel Institute of Technology, Haifa, Israel. Dagan, Ido; Justeson, John; Lappin, Shalom; Leass, Herbert; and Ribak, Amnon (in 27 Bosch (1988) suggests a psychological processing model in which hearers rely on first pass syntactically based strategies for initial linking of pronouns to antecedent NPs. press). "Syntax and lexical statistics in anaphora resolution." Applied Artificial Intelligence. find
  7. Grosz, Barbara; Joshi, Aravind; and Weinstein, Scott (1983). "Providing a unified account of definite noun phrases in discourse." In Proceedings, 21st Annual Meeting of the Association of Computational Linguistics, 44-50. find
  8. Grosz, Barbara; Joshi, Aravind; and Weinstein, Scott (1986, unpublished). "Towards a computational theory of discourse interpretation." Harvard University and University of Pennsylvania. find
  9. Grosz, Barbara, and Sidner, Candice (1986). "Attention, intentions, and the structure of discourse." Computational Linguistics 12:175-204. find our card
  10. Guenthner, Franz, and Lehmann, Hubert (1983). "Rules for pronominalization." In Proceedings, First Annual Meeting of the European Chapter of the ACL, 144-151. find
  11. Hobbs, Jerry (1978). "Resolving pronoun references." Lingua 44:311-338. find our card
  12. Johnson, David (1977). "On relational constraints on grammars." In Syntax and Semantics 8, edited by P. Cole and J. Sadock, 151-178. New York: Academic Press. find
  13. Kamp, Hans (1981). "A theory of truth and semantic representation." In Formal Methods in the Study of Language, edited by J. Groenendijk, T. Janssen, and M. Stokhof. Amsterdam: Mathematisch Centrum Tracts. find
  14. Keenan, Edward, and Comrie, Bernard (1977). "Noun phrase accessibility and universal grammar." Linguistic Inquiry 8:62-100. find
  15. Lappin, Shalom (1985). "Pronominal binding and coreference." Theoretical Linguistics 12:241-263. find
  16. Lappin, Shalom, and McCord, Michael (1990a). "A syntactic filter on pronominal anaphora in slot grammar." In Proceedings, 28th Annual Meeting of the Association for Computational Linguistics, 135-142. find
  17. Lappin, Shalom, and McCord, Michael (1990b). "Anaphora resolution in slot grammar." Computational Linguistics 16:197-212. find
  18. Leass, Herbert, and Schwall, Ulrike (1991). "An anaphora resolution procedure for machine translation." IWBS Report 172, IBM Germany Scientific Center, Heidelberg, Germany. find
  19. McCord, Michael (1989a). "Design of LMT: A prolog-based machine translation system." Computational Linguistics 15:33-52. find our card
  20. McCord, Michael (1989b). "A new version of the machine translation system LMT." Literary and Linguistic Computing 4:218-229. find
  21. McCord, Michael (1990). "Slot grammar: A system for simpler construction of practical natural language grammars." In Natural Language and Logic: International Scientific Symposium, edited by R. Studer, 118-145. Lecture Notes in Computer Science, Berlin: Springer Verlag. find
  22. McCord, Michael (1993). "Heuristics for broad-coverage natural language parsing." In Proceedings, ARPA Human Language Technology Workshop, University of Pennsylvania. find
  23. McCord, Michael (in press). "The slot grammar system." In Unification in Grammar, edited by Jiirgen Wedekin and Christian Rohrer (also IBM Research Report RC 17313). Cambridge, MA: MIT Press. find
  24. McCord, Michael; Bernth, Arendse; Lappin, Shalom; and Zadrozny, Wlodek (1992). "Natural language processing within a slot grammar framework." International Journal on Artificial Intelligence Tools 1:229-277. find
  25. Rich, Elaine, and LuperFoy, Susann (1988). "An architecture for anaphora resolution." In Proccedings, ACL Conference on Applied Natural Language Processing, 18-24. find
  26. Sidner, Candice (1981). "Focusing for interpretation of pronouns." American Journal of Computational Linguistics 7:217-231. find
  27. Sidner, Candice (1983). "Focusing in the comprehension of definite anaphora." In Computational Models of Discourse, edited by Michael Brady and Robert Berwick, 267-330. Cambridge, MA: MIT Press. find
  28. Walker, Marilyn (1989). "Evaluating discourse processing algorithms." In Proceedings, 27th Annual Meeting of the Association for Computational Linguistics, 251-261. find
  29. Walker, Marilyn; Iida, Masayo; and Cote, Sharon (1990). "Centering in Japanese discourse." In Proceedings, 13th International Conference on Computational Linguistics, 1-6. find
  30. Webber, Bonnie (1988). "Discourse Deixis: Reference to discourse segments." In Proceedings, 26th Annual Meeting of the Association for Computational Linguistics, 113-121. find
  31. Williams, Edwin (1984). "Grammatical relations." Linguistic Inquiry 15:639~73. find