Resolving Discourse Deictic Anaphora in Dialogues

What it is about

The paper addresses anaphora in spoken dialogue, using five Switchboard telephone conversations (three for testing). Only 45% of anaphors in that corpus have a noun-phrase antecedent; about 23% are discourse deictic (they refer to a clause, a verb phrase or a stretch of talk), about 19% are inferrable-evoked plural pronouns, and about 13% are vague. The method first classifies each pronoun or demonstrative as individual, discourse deictic or vague, using the type of the predicate it sits under (propositional-attitude verbs, copulas such as "is true", and "do"). It then resolves it against a ranked context: a list of abstract referents, the clause to the left, the previous turn, and the previous sentence. Dialogue is split into initiations and acknowledgements, and an entity enters the discourse model only when the other speaker acknowledges it. Precision on abstract anaphora is 63.6% and recall 70%, and the authors say precision is low because the classification step is weak.

Ideas

• I1. Classify before resolving. In coref, assign an anaphor type from the UD structure first: a pronoun under a clausal complement of a verb like "know", or under a copula with "true" or "right", skips noun-phrase candidates and goes to clause candidates. Measure the type classification on a dialogue set we may use, and resolve only after the type is fixed.
• I2. Acknowledgement gate for dialogue. In the dialogue lessons (chapter 38), an entity from an utterance that gets no acknowledgement stays out of the candidate list for the next turn. Test on our own dialogue lessons with and without the gate, two identical runs, two-fold cv.
• I3. A ranked context list with the slot named in the output: abstract list, same clause to the left, previous clause, previous sentence. The explain output then says which slot won, which is what Mova already aims for.
• I4. Build verb-type lists for Mova from our own lexicon: which verbs take a sentential complement, which copula adjectives apply only to propositions ("true", "wrong", "right"). The paper says such lists are the weak part; use them as data in en, with a source note for each entry.

For Mova

The chapters are 35. Reference and coreference and 38. Dialogue, discourse markers and reported speech. The code is coref/src/mention.rs and coref/src/sieve.rs, plus the verb lists in en. The Switchboard corpus is LDC-licensed: tests only, and never in a release. Build any dialogue test set from our own or open material.

Topics: Discourse analysis · Coreference resolution

Cite

The paper itself, not this card. BibTeX from the library's record of the work. DOI 10.3115/977035.977042

@inproceedings{eckert1999resolving,
  title = {{Resolving Discourse Deictic Anaphora in Dialogues}},
  author = {Eckert, Miriam and Strube, Michael},
  booktitle = {Ninth Conference of the European Chapter of the Association for Computational Linguistics},
  year = {1999},
  doi = {10.3115/977035.977042},
  url = {https://doi.org/10.3115/977035.977042}
}

References

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

1. Nicholas Asher. 1993. Reference to Abstract Objects. Kluwer, Dordrecht. find
2. Adriana Belletti and Luigi Rizzi. 1988. Psych verbs and theta theory. Natural Language and Linguistic Theory, 6:291-352. find
3. Donna Byron and Amanda Stent. 1998. A preliminary model of centering in dialog. In Proceedings of the 17 th International Conference on Computational Linguistics and 36 th Annual Meeting of the Association for Computational Linguistics, Montrral, Qurbec, Canada, 10-14 August 1998, pages 1475-1477. find
4. Jean Carletta, Amy Isard, Stephen Isard, Jacqueline Kowtko, Gwyneth Doherty-Sneddon, and Anne Anderson. 1997. The reliability of a dialogue structure coding scheme. Computational Linguistics, 23(1):13-31. find
5. Jean Carletta. 1996. Assessing agreement on classification tasks: The kappa statistic. Computational Linguistics, 22(2):249-254. find
6. Herbert H. Clark and Edward F. Schaefer. 1989. Contributing to discourse. Cognitive Science, 13:259-294. find
7. Ido Dagan and Alon Itai. 1991. A statistical filter for resolving pronoun references. In Y.A. Feldman and A. Bruckstein, editors, Artificial Intelligence and Computer Vision, pages 125-135. Elsevier, Amsterdam. t3sten Dahl and Christina Hellman. 1995. What happens when we use an anaphor. In Presentation at the XVth Scandinavian Conference of Linguistics Oslo, Norway. find
8. Susan Garnsey, Neal Pearlmutter, Elizabeth Myers, and Melanie Lotocky. 1997. Contributions of verb bias and plausibility to the comprehension of temporarily ambiguous sentences. Journal of Memory and Language, 37:58-93. find
9. Barbara J. Grosz, Aravind K. Joshi, and Scott Weinstein. 1995. Centering: A framework for modeling the local coherence of discourse. Computational Linguistics, 21 (2):203-225. find our card
10. Jeanette K. Gundel, Nancy Hedberg, and Ron Zacharski. 1993. Cognitive status and the form of referring expressions in discourse. Language, 69:274-307. find
11. Osvaldo Jaeggli. 1986. Arbitrary plural pronominals. Natural Language and Linguistic Theory, 4:43-76. find
12. LDC. 1993. Switchboard. Linguistic Data Consortium. University of Pennsylvania, Philadelphia, Penn. find
13. Susann LuperFoy. 1992. The representation of multimodal user interface dialogues using discourse pegs. In Proceedings of the 3 0 th Annual Meeting of the Association for Computational Linguistics, Newark, Del., 28 June - 2 July 1992, pages 22-31. find
14. Rebecca Passonneau and Diane Litman. 1997. Discourse segmentation by human and automated means. Computational Linguistics, 23(1): 103-139. find
15. Paul Postal and Geoffrey Pullum. 1988. Expletive noun phrases in subcategorized positions. Linguistic Inquiry, 19:635-670. find
16. Michael Strube. 1998. Never look back: An alternative to centering. In Proceedings of the 17 th International Conference on Computational Linguistics and 36 th Annual Meeting of the Association for Computational Linguistics, Montrfal, Qurbec, Canada, 10-14 August 1998, pages 1251-1257. 1994. A Computational Theory of Grounding in Natural Language Conversation. Ph:D. thesis, Department of Computer Sci-David R. Traum. ence, University of Rochester. find
17. Marilyn A. Walker. 1998. Centering, anaphora resolution, and discourse structure. In M.A. Walker, A.K. Joshi, and E.E Prince, editors, Centering Theory in Discourse, pages 401-435. Oxford University Press, Oxford, U.K. find
18. Bonnie L. Webber. 1991. Structure and ostension in the interpretation of discourse deixis. Language and Cognitive Processes, 6(2): 107-135. find

This page in Markdown → https://md.mova.elamur.ai/free/en/docs/papers/prag-1999-eckert-resolving-discourse-deictic-anaphora-dialogues/