Elamur Research⌕
☰
home / papers / 2006 / Automated Multiword Expression Prediction for Grammar Engineering

Automated Multiword Expression Prediction for Grammar Engineering

Paper · 2006 · Yi Zhang, Valia Kordoni, Aline Villavicencio et al.

Summary

Even large hand-built grammars do not cover all words and constructions. Multiword expressions (MWEs), such as phrasal verbs, compound nouns and fixed phrases, are a constant cause of parse failures. The authors propose a semi-automatic way to extend the lexicon of the HPSG grammar English Resource Grammar (ERG). First, the ERG together with the PET parser is run on approximately 1.8 million short sentences from the BNC. Then the error mining method proposed by van Noord is used to find n-grams that occur almost exclusively in sentences without a parse. Next, the candidates are validated using the World Wide Web as a corpus together with statistical measures. This filters out noise, such as typographical errors. For the remaining candidates, a maximum entropy classifier trained on the Redwoods Treebank determines the most probable atomic lexical type, taking context into account. The resulting entries are added to the grammar. As a result, coverage on sentences containing MWEs increased by 14.4%. The method does not require electronic dictionaries or ontologies. It can therefore be applied to other languages and grammars, including low-resource ones.

Key points

  • Data: a subset of the written part of BNC 2.0, consisting of sentences of up to 20 words containing only ASCII characters, approximately 1.8 million unique sentences. Grammar: ERG (January 2006 release); parser: PET.
  • Distribution of parsing outcomes: 35.80% of sentences received a parse (P), 53.82% contain unknown words (L), and for 10.38% no parse was found (N).
  • Error mining used the N and P sentences with a frequency threshold of 5. This yielded scores for about 35K n-grams. A parsability score below 0.1 was obtained by 798 unigrams, 2,011 bigrams and 937 trigrams.
  • 311 MWE candidates passed the web-based validation. Across 6,248 BNC sentences, the predictor generated 1,936 distinct lexical entries. Only those occurring at least 5 times were added to the grammar, giving 373 new MWE entries.
  • Coverage increased by 14.4%, but on this difficult set of sentences it still remains noticeably lower than on the corpus as a whole. Hence, not all MWEs can be handled simply by adding new lexical entries.

Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.

 Institutions

🎓Saarland UniversitySaarbrücken · Germany148 works · 18,287 citations🎓Universidade Federal do Rio Grande do SulPorto Alegre · Brazil8 works · 522 citations

 Fields, topics and concepts

📚 Shelves 4The field of science this work belongs to (OpenAlex fields).Computer Science · 16,667 worksArtificial Intelligence · 16,194 worksArts and Humanities · 16,039 worksLanguage and Linguistics · 15,891 works

🧭 Topics 2Research subjects and languages the work is about.Syntax & grammar formalisms · 214 worksIdioms & multiword expressions · 217 works

Cite

Cite as Yi Zhang, Valia Kordoni, Aline Villavicencio and Marco Idiart. 2006. Automated Multiword Expression Prediction for Grammar Engineering. Proceedings of the Workshop on Multiword Expressions: Identifying and Exploiting Underlying Properties.

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

@inproceedings{zhang2006automated,
  title = {{Automated Multiword Expression Prediction for Grammar Engineering}},
  author = {Yi Zhang, Valia Kordoni, Aline Villavicencio and Marco Idiart},
  booktitle = {Proceedings of the Workshop on Multiword Expressions: Identifying and Exploiting Underlying Properties},
  year = {2006},
  doi = {10.3115/1613692.1613700},
  url = {https://doi.org/10.3115/1613692.1613700}
}

References

21 references, 4 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Timothy Baldwin, Emily M. Bender, Dan Flickinger, Ara Kim, and Stephan Oepen. 2004. Road-testing the English Resource Grammar over the British National Corpus. In Proceedings of the Fourth International Conference on Language Resources and Evaluation (LREC 2004), Lisbon, Portugal.
  2. Timothy Baldwin. 2005. Bootstrapping deep lexical resources: Resources for courses. In Proceedings of the ACL-SIGLEX Workshop on Deep Lexical Acquisition, pages 67–76, Ann Arbor, Michigan, June. Association for Computational Linguistics. find
  3. Gosse Bouma and Begońa Villada. 2002. Corpus-based acquisition of collocational prepositional phrases. In Proceedings of the Computational Linguistics in the Netherlands (CLIN) 2001, University of Twente. find
  4. Gosse Bouma, Gertjan van Noord, and Robert Malouf. 2001. Alpino: Wide-coverage computational analysis of dutch. In Computational Linguistics in The Netherlands 2000. find
  5. Entries Added 0 373 ERG ERG+MWE(Web) Table 6: Parser coverage on “difficult” sentences before/after adding MWE lexical entries Lou Burnard. 2000. User Reference Guide for the British National Corpus. Technical report, Oxford University Computing Services. find
  6. M. Butt, S. Dipper, A. Frank, and T.H. King. 1999. Writing large-scale parallel grammars for english, french, and german. In Proceedings of the LFG99 Conference. CSLI Publications. find
  7. Ulrich Callmeier. 2000. PET – a platform for experimentation with efficient HPSG processing techniques. Journal of Natural Language Engineering, 6(1):99–108. find
  8. Bob Carpenter. 1992. The Logic of Typed Feature Structures. Cambridge University Press, Cambridge, England. find
  9. Dan Flickinger. 2000. On building a more efficient grammar by exploiting types. Natural Language Engineering, 6(1):15–28.
  10. Gregory Grefenstette. 1999. The World Wide Web as a resource for example-based machine translation tasks. In Proceedings of ASLIB, Conference on Translating and the Computer, London. find
  11. Ray Jackendoff. 1997. Twistin’ the night away. Language, 73:534–59. find
  12. Frank Keller, Maria Lapata, and Olga Ourioupina. 2002. Using the Web to overcome data sparseness. In Jan Hajič and Yuji Matsumoto, editors, Proceedings of the Conference on Empirical Methods in Natural Language Processing, pages 230–237, Philadelphia. find
  13. Adam Kilgarriff and Gregory Grefenstette. 2003. Introduction to the special issue on web as corpus. Computational Linguistics, 29. find
  14. Robert Malouf. 2002. A comparison of algorithms for maximum entropy parameter estimation. In Proceedings of the Sixth Conferencde on Natural Language Learning (CoNLL-2002), pages 49–55.
  15. Stephan Oepen and John Carroll. 2000. Ambiguity packing in constraint-based parsing — practical results. In Proceedings of the 1st Conference of the North American Chapter of the ACL, pages 162– 169, Seattle, WA. find
  16. Carl J. Pollard and Ivan A. Sag. 1994. Head-Driven Phrase Structure Grammar. University of Chicago Press, Chicago, Illinois. find
  17. Item \# 6246 6246 Covered \# 268 1168 Coverage 4.3% 18.7% Ivan Sag, Timothy Baldwin, Francis Bond, Ann Copestake, and Dan Flickinger. 2002. Multiword expressions: A pain in the neck for NLP. In Proceedings of the 3rd International Conference on Intelligent Text Processing and Computational Linguistics…
  18. Gertjan van Noord. 2004. Error mining for wide-coverage grammar engineering. In Proceedings of the 42nd Meeting of the Association for Computational Linguistics (ACL’04), Main Volume, pages 446–453, Barcelona, Spain, July. find
  19. Aline Villavicencio, Francis Bond, Anna Korhonen, and Diana McCarthy. 2005. Introduction to the special issue on multiword expressions: having a crack at a hard nut. Journal of Computer Speech and Language Processing, 19. find
  20. Aline Villavicencio. 2005. The availability of verb-particle constructions in lexical resources: How much is enough? Journal of Computer Speech and Language Processing, 19. find
  21. Yi Zhang and Valia Kordoni. 2006. Automated deep lexical acquisition for robust open texts processing. In Proceedings of the Fifth International Conference on Language Resources and Evaluation (LREC 2006), Genoa, Italy. find

Updated

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/zhang-2006-automated-multiword-expression-prediction-grammar/