2005 ComparingKnowSourcesForNomAnaphoraRes

From GM-RKB
Jump to navigation Jump to search

Subject Headings: Anaphora Resolution Algorithm, Nominal Anaphora, BNC Corpus, WSJ Corpus, WordNet, World Wide Web.

Notes

  • It makes use of manually constructed Ontologies.
  • It makes use of the World Wide Web.
  • It confirms that a significant proportion of Lexical Links often exploited in Coreference are not included in WordNet.
  • It suggests that there is benefit to be drawn from using the large but noisy Web (over just a small and clean BNC-based one).

Cited By

~42 http://scholar.google.com/scholar?cites=8240602663420466578

2007

Quotes

Abstract

We compare two ways of obtaining lexical knowledge for antecedent selection in other-anaphora and definite noun phrase coreference. Specifically, we compare an algorithm that relies on links encoded in the manually created lexical hierarchy WordNet and an algorithm that mines corpora by means of shallow lexico-semantic patterns. As corpora we use the British National Corpus (BNC), as well as the Web, which has not been previously used for this task. Our results show that (a) the knowledge encoded in WordNet is often insufficient, especially for anaphor' antecedent relations that exploit subjective or context-dependent knowledge; (b) for other-anaphora, the Web-based method outperforms the WordNet-based method; (c) for definite NP coreference, the Web-based method yields results comparable to those obtained using WordNet over the whole data set and outperforms the WordNet-based method on subsets of the data set; (d) in both case studies, the BNC-based method is worse than the other methods because of data sparseness. Thus, in our studies, the Web-based method alleviated the lexical knowledge gap often encountered in anaphora resolution and handled examples with context-dependent relations between anaphor and antecedent. Because it is inexpensive and needs no hand-modeling of lexical knowledge, it is a promising knowledge source to integrate into anaphora resolution systems.

1. Introduction

Most work on anaphora resolution has focused on pronominal anaphora, often achieving good accuracy. Kennedy and Boguraev (1996), Mitkov (1998), and Strube, Rapp, and Mueller (2002), for example, report accuracies of 75.0%, 89.7%, and an F-measure of 82.8% for personal pronouns, respectively. Less attention has been paid to nominal anaphors with full lexical heads, which cover a variety of phenomena, such as coreference (Example (1)), bridging (Clark 1975; Example (2)), and comparative anaphora (Examples (3–4)). [1 In all examples presented in this article, the anaphor is typed in boldface and the correct antecedent in italics. The abbreviation in parentheses at the end of each example specifies the corpus from which the example is taken: WSJ stands for the Wall Street Journal, Penn Treebank, release 2; BNC stands for British National Corpus (Burnard 1995), and MUC-6 for the combined training/test set for the coreference task of the Sixth Message Understanding Conference (Hirschman and Chinchor 1997).]

  1. The death of Maxwell, the British publishing magnate whose empire collapsed in ruins of fraud, and who was the magazine’s publisher, gave the periodical a brief international fame. (BNC)
  2. [. . .] you don’t have to undo the jacket to get to the map — particularly important when it’s blowing a hooley. There are elasticated adjustable drawcords on the hem, waist and on the hood. (BNC)
  3. In addition to increasing costs as a result of greater financial exposure for members, these measures could have other, far-reaching repercussions. (WSJ)
  4. The ordinance, in Moon Township, prohibits locating a group home for the handicapped within a mile of another such facility. (WSJ)

In Example (1), the definite noun phrase (NP) the periodical corefers with the magazine. [2 In this article, we restrict the notion of definite NPs to NPs modified by the article ‘the.’] In Example (2), the definite NP the hood can be felicitously used because a related entity has already been introduced by the NP the jacket, and a part-of relation between the two entities can be established. Examples (3)–(4) are instances of other-anaphora. Other-anaphora are a subclass of comparative anaphora (Halliday and Hasan 1976; Webber et al. 2003) in which the anaphoric NP is introduced by a lexical modifier (such as other, such, and comparative adjectives) that specifies the relationship (such as set-complement, similarity and comparison) between the entities invoked by anaphor and antecedent. For other-anaphora, the modifiers other or another provide a set-complement to an entity already evoked in the discourse model. In Example (3), the NP other, far-reaching repercussions refers to a set of repercussions excluding increasing costs and can be paraphrased as other (far-reaching) repercussions than (increasing) costs. Similarly, in Example (4), the NP another such facility refers to a group home which is not identical to the specific (planned) group home mentioned before.

A large and diverse amount of lexical or world knowledge is usually necessary to understand anaphors with full lexical heads. For the examples above, we need the knowledge that magazines are periodicals, that hoods are parts of jackets, that costs can be or can be viewed as repercussions of an event, and that institutional homes are facilities. Therefore, many resolution systems that handle these phenomena (Vieira and Poesio 2000; Harabagiu, Bunescu, and Maiorano 2001; Ng and Cardie 2002b; Modjeska 2002; Gardent, Manuelian, and Kow 2003, among others) rely on hand-crafted resources of lexico-semantic knowledge, such as the WordNet lexical hierarchy (Fellbaum 1998).3 In Section 2, we summarize previous work that has given strong indications that such resources are insufficient for the entire range of full NP anaphora. Additionally, we discuss some serious methodological problems that arise when fixed ontologies are used that have been encountered by previous researchers and/or us: the costs of building, maintaining and mining ontologies; domain-specific and context-dependent knowledge; different ways of encoding information; and sense ambiguity.

7. Conclusions

We have explored two different ways of exploiting lexical knowledge for antecedent selection in other-anaphora and definite NP coreference. Specifically, we have compared a hand-crafted and -structured source of information such as WordNet and a simple and inexpensive pattern-based method operating on corpora. As corpora we have used the BNC and also suggested the Web as the biggest corpus available.

We confirmed results by other researchers that show that a substantial number of lexical links often exploited in coreference are not included in WordNet. We have also shown the presence of an even more severe knowledge gap for other-anaphora (see also Question 1 in Section 1). Largely because of this knowledge gap, the novel Web-based method that we proposed proved better than WordNet at resolving other-anaphora. Although the gains for coreference are not as high, the Web-based method improves more substantially on string-matching techniques for coreference than WordNet does (see the success rate beyond StrSetv2n for coreference, Section 5.5). In both studies, the Web-based method clearly outperformed the BNC-based one. This shows that, for our tasks, overcoming data sparseness was more important than working with a manually controlled, virtually noise-free, but relatively small corpus, which addresses Question 2 in Section 1: Corpus-induced knowledge can indeed rival and even outperform the knowledge obtained via lexical hierarchies, as long as the corpus is large enough. Corpus-based methods can therefore be a very useful complement to resolution algorithms for languages for which hand-crafted taxonomies have not yet been created but for which large corpora do exist. In answer to Question 3 in Section 1, our results suggest that different anaphoric phenomena suffer in varying degrees from missing knowledge and that the Web-based method performs best when used to deal with phenomena that standard taxonomy links do not capture that easily or that frequently exploit subjective and context-dependent knowledge.

In addition, the Web-based method that we propose does not suffer from some of the intrinsic limitations of ontologies, specifically, the problem of what knowledge should be included (see Section 2.2). It is also inexpensive and does not need any postprocessing of the Web pages returned or any hand-modeling of lexical knowledge.

To summarize, antecedent selection for other-anaphora and definite NP coreference without hand-crafted lexical knowledge is feasible. This might also be the case for yet other full NP anaphora types with similar properties — an issue that we will explore in future work.

References

  • Ariel, Mira. (1990). Accessing Noun Phrase Antecedents. Routledge, London and New York.
  • Berland, Matthew and Eugene Charniak. (1999). Finding parts in very large corpora. In: Proceedings of the 37th Annual Meeting of the Association for Computational Linguistics, pages 57–64, Providence, RI.
  • Bierner, Gann. (2001). Alternative phrases and natural language information retrieval. In: Proceedings of the 39th Annual Meeting of the Association for Computational Linguistics, pages 58–65, Toulouse, France.
  • Burnard, Lou, (1995). Users’ Reference Guide, British National Corpus. British National Corpus Consortium, Oxford.
  • Caraballo, Sharon. (1999). Automatic acquisition of a hypernym-labelled noun hierarchy from text. In: Proceedings of the 37th Annual Meeting of the Association for Computational Linguistics, pages 120–126, Providence, RI.
  • Chinchor, Nancy. (1997). MUC-7 named entity task definition. In: Proceedings of the Seventh Conference on Message Understanding, Washington, DC. 399 Computational Linguistics Volume 31, Number 3.
  • Christ, Oliver, (1995). The XKWIC User Manual. Institute for Computational Linguistics, University of Stuttgart.
  • Clark, Herbert H. 1975. Bridging. In: Proceedings of the Conference on Theoretical Issues in Natural Language Processing, pages 169–174, Cambridge, MA.
  • Connolly, Dennis, John D. Burger, and David S. Day. (1997). A machine learning approach to anaphoric reference. In Daniel Jones and Harold Somers, editors, New Methods in Language Processing.
  • University College London Press, London, pages 133–144.
  • Curran, James and Stephen Clark. (2003). Language independent NER using a maximum entropy tagger. In: Proceedings of the Seventh Conference on Natural Language Learning (CoNLLO3), pages 164–167, Edmonton, Alberta, Canada.
  • Fellbaum, Christiane, editor. (1998). WordNet: An Electronic Lexical Database. MIT Press, Cambridge, MA.
  • Fraurud, Kari. (1990). Definiteness and the processing of NPs in natural discourse.
  • Journal of Semantics, 7:395–433.
  • Gardent, Claire, Helene Manuelian, and Eric Kow. (2003). Which bridges for bridging definite descriptions? In: Proceedings of the EACL 2003 Workshop on Linguistically Interpreted Corpora, pages 69–76, Budapest.
  • Grefenstette, Gregory. (1999). The WWW as a resource for example-based MT tasks. In: Proceedings of ASLIB’99: Translating and the Computer 21, London.
  • Gundel, Jeanette, Nancy Hedberg, and Ron Zacharski. (1993).
  • Cognitive status and the form of referring expressions in discourse. Language, 69(2):274–307.
  • Hahn, Udo, Michael Strube, and Katja Markert. (1996).
  • Bridging textual ellipses. In: Proceedings of the 16th International Conference on Computational Linguistics, pages 496–501, Copenhagen.
  • Halliday, Michael A. K. and Ruqaiya Hasan. 1976.
  • Cohesion in English. Longman,London.
  • Sanda M. Harabagiu. (1997). WordNet-based Inference of Textual Context, Cohesion and Coherence. Ph.D. thesis, University of Southern California.
  • Sanda M. Harabagiu, Razvan C. Bunescu, and Steven J. Maiorano. (2001). Text and knowledge mining for coreference resolution. In: Proceedings of the Second Conference of the North American Chapter of the ACL, pages 55–62, Pittsburgh.
  • Hawkins, John A. 1978. Definiteness and Indefiniteness. Croom Helm, London.
  • Hearst, Marti. (1992). Automatic acquisition of hyponyms from large text corpora. In: Proceedings of the 14th International Conference on Computational Linguistics, Nantes, France.
  • Hirschman, Lynette and Nancy Chinchor. (1997).
  • MUC-7 coreference task definition. In: Proceedings of the Seventh Conference on Message Understanding,Washington, DC.
  • Humphreys, Kevin, Robert Gaizauskas, Saliha Azzam, Chris Huyck, Brian Mitchell, and Hamish Cunningham. 1997.
  • University of Sheffield: Description of the LaSie-II system as used for MUC-7. In: Proceedings of the Seventh Message Understanding Conference (MUC-7), Washington, DC.
  • Kameyama, Megumi. (1997). Recognizing referential links: An information extraction perspective. In: Proceedings of the ACL-1997 Workshop on Operational Factors in Practical, Robust Anaphora Resolution for Unrestricted Texts, pages 46–53, Madrid.
  • Keller, Frank and Maria Lapata. (2003). Using theWeb to obtain frequencies for unseen bigrams. Computational Linguistics, 29(3):459–484.
  • Kennedy, Christopher and Branimir Boguraev. (1996). Anaphora for everyone: Pronominal anaphora resolution without a parser. In: Proceedings of the 16th International Conference on Computational Linguistics, pages 113–118, Copenhagen.
  • Markert, Katja, Malvina Nissim, and Natalia N. Modjeska. (2003). Using the Web for nominal anaphora resolution. In Robert Dale, Kees van Deemter, and Ruslan Mitkov, editors, Proceedings of the EACL Workshop on the Computational Treatment of Anaphora, pages 39–46, Budapest.
  • McCoy, Kathleen and Michael Strube. 1999.
  • Generating anaphoric expressions: Pronoun or definite description? In ACL-99 Workshop on the Relation of Discourse/Dialogue Structure and Reference, pages 63–71, College Park, MD.
  • Meyer, Ingrid. (2001). Extracting knowledge-rich contexts for terminography. In Didier Bourigault, Christian Jacquemin, and Marie-Claude L’Homme, editors, Recent Advances in Computational Terminology. John Benjamins, Amsterdam, pages 279–301.
  • Meyer, Josef and Robert Dale. (2002). Mining a corpus to support associative anaphora resolution. In: Proceedings of 400 Markert and Nissim Knowledge Sources for Anaphora Resolution the Fourth International Conference on Discourse Anaphora and Anaphor Resolution, Lisbon.
  • (Mitkov, 1998) ⇒ Ruslan Mitkov. (1998). “Robust Pronoun Resolution With Limited Knowledge.” In: Proceedings of the 17th International Conference on Computational Linguistics and 36th Annual Meeting of the Association for Computational Linguistics (ACL 1998).
  • Modjeska, Natalia N. (2002). Lexical and grammatical role constraints in resolving other-anaphora. In: Proceedings of DAARC 2002, pages 129–134, Lisbon.
  • Modjeska, Natalia N. (2003). Resolving other-anaphora. Ph.D. thesis, School of Informatics, University of Edinburgh.
  • Modjeska, Natalia N., Katja Markert, and Malvina Nissim. (2003). Using theWeb in machine learning for other-anaphora resolution. In: Proceedings of the 2003 Conference on Empirical Methods in Natural Language Processing, pages 176–183, Sapporo, Japan.
  • Ng, Vincent. (2004). Learning noun phrase anaphoricity to improve coreference resolution: Issues in representation and optimization. In: Proceedings of the 42nd Annual Meeting of the Association for Computational Linguistics, pages 151–158, Barcelona.
  • Ng, Vincent and Claire Cardie. 2002a. Identifying anaphoric and non-anaphoric noun phrases to improve coreference resolution. In: Proceedings of the 19th International Conference on Computational Linguistics, pages 730–736, Taipei.
  • Ng, Vincent and Claire Cardie. 2002b. Improving machine learning approaches to coreference resolution. In: Proceedings of the 40th Annual Meeting of the Association for Computational Linguistics, pages 104–111, Philadelphia.
  • Poesio, Massimo, Tomonori Ishikawa, Sabine Schulte imWalde, and Renata Vieira. (2002). Acquiring lexical knowledge for anaphora resolution. In: Proceedings of the Third International Conference on Language Resources and Evaluation, pages 1220–1224, Las Palmas, Canary Islands.
  • Poesio, Massimo, Rahul Mehta, Axel Maroudas, and Janet Hitzeman. 2004.
  • Learning to resolve bridging references. In: Proceedings of the 42nd Annual Meeting of the Association for Computational Linguistics, pages 143–150, Barcelona.
  • Poesio, Massimo, Renata Vieira, and Simone Teufel. (1997). Resolving bridging references in unrestricted text. In Ruslan Mitkov, editor, Proceedings of the ACL Workshop on Operational Factors in Robust Anaphora Resolution, pages 1–6, Madrid.
  • Preiss, Judita. (2002). Anaphora resolution with word sense disambiguation. In: Proceedings of SENSEVAL-2, pages 143–146, Philadelphia.
  • Preiss, Judita, Caroline Gasperin, and Ted Briscoe. (2004). Can anaphoric definite descriptions be replaced by pronouns? In: Proceedings of the Fourth International Conference on Language Resources and Evaluation, pages 1499–1502, Lisbon.
  • Shinzato, Keiji and Kentaro Torisawa. (2004). Acquiring hyponymy relations fromWeb documents. In: Proceedings of the Conference of the North American Chapter of the ACL, pages 73–80, Boston.
  • Soon,Wee Meng, Hwee Tou Ng Ng, and Daniel Chung Yung Lim. (2001). A machine learning approach to coreference resolution of noun phrases. Computational Linguistics, 27(4):521–544.
  • Michael Strube, Stefan Rapp, and Christoph Mueller. (2002). The influence of minimum edit distance on reference resolution. In: Proceedings of the 2002 Conference on Empirical Methods in Natural Language Processing, pages 312–319, Philadelphia.
  • Uryupina, Olga. (2003). High-precision identification of discourse new and unique noun phrases. In: Proceedings of the ACL 2003 Student Workshop, pages 80–86, Sapporo, Japan.
  • van Deemter, Kees and Rodger Kibble. (2000). On coreferring: Coreference in MUC and related annotation schemes. Computational Linguistics, 26(4):615–662.
  • Vieira, Renata and Massimo Poesio. (2000). An empirically-based system for processing definite descriptions. Computational Linguistics, 26(4):539–593.
  • Webber, Bonnie, Matthew Stone, Aravind Joshi, and Alistair Knott. (2003). Anaphora and discourse structure. Computational Linguistics, 29(4):545–587.
  • Yang, Xiaofeng, Guodong Zhou, Jian Su, and Chew Lim Tan. (2003). Coreference resolution using competition learning approach. In: Proceedings of the 41st Annual Meeting of the Association for Computational Linguistics, pages 176–183, Sapporo, Japan.

,

 AuthorvolumeDate ValuetitletypejournaltitleUrldoinoteyear
2005 ComparingKnowSourcesForNomAnaphoraResKatja Markert
Malvina Nissim		Comparing Knowledge Sources for Nominal Anaphora Resolutionhttp://www.mitpressjournals.org/doi/abs/10.1162/089120105774321064
Retrieved from "http://www.gabormelli.com/RKB/index.php?title=2005_ComparingKnowSourcesForNomAnaphoraRes&oldid=639381"