Elamur Research⌕
☰
home / papers / 2007 / Pulling their Weight: Exploiting Syntactic Forms for the Automatic Identification of Idiomatic Expressions in Context

Pulling their Weight: Exploiting Syntactic Forms for the Automatic Identification of Idiomatic Expressions in Context

Summary

The paper tackles token-level idiom identification: deciding whether a specific occurrence of an ambiguous verb+noun combination (VNC), such as make a face, is meant idiomatically or literally. Earlier work mostly classified expression types, or treated token classification as sense disambiguation based only on surrounding words. The authors instead use a type-level property, syntactic fixedness. Idioms tend to appear in one or a few canonical forms (for example, a particular voice, determiner and noun number), while literal uses vary more. They propose three unsupervised classifiers. CForm labels a token idiomatic whenever it appears in an automatically learned canonical form. Two Diff methods compare a token's co-occurrence vector against vectors representing the idiomatic and literal readings. The data are VNCs and canonical forms from Fazly and Stevenson (2006), with annotated sentences drawn from the BNC. All three methods beat the majority-class baseline, and the best one comes close to a supervised kNN classifier. This shows that type-based syntactic knowledge can stand in for labelled training data.

Key points

  • The candidate set was 60 VNCs, each with BNC frequency ≥ 20 and listed in an idiom dictionary. The authors annotated 1180 tokens of 28 VNCs, and about 40% of these tokens were literal uses.
  • Canonical forms come from the unsupervised FS06 method, which selects forms occurring much more often than the expression's average form.
  • The two Diff variants estimate the literal meaning differently. Diff(I-CF, L-NCF) builds it from non-canonical-form usages. Diff(I-CF, L-Comp) builds it compositionally by summing the component word vectors. Similarity is measured with cosine.
  • Overall, the best unsupervised accuracy was 72.4%, compared with 76.2% for a standard supervised approach.
  • On VNCs with mostly idiomatic usage, CForm reached 84.7%, against 82.3% for 5NN and an 81.4% baseline. On VNCs where idiomatic use is less frequent, the unsupervised methods fell behind 5NN (72.4%), which the authors attribute to less reliable canonical forms.

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

Cited by

42 papers of our library cite it (of the 140673 whose bibliographies we read); 8 of them have a page here, 34 not yet.

Cite

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

@inproceedings{cook2007pulling,
  title = {{Pulling their Weight: Exploiting Syntactic Forms for the Automatic Identification of Idiomatic Expressions in Context}},
  author = {Cook, Paul and Fazly, Afsaneh and Stevenson, Suzanne},
  booktitle = {Proceedings of the Workshop on A Broader Perspective on Multiword Expressions},
  year = {2007},
  url = {https://aclanthology.org/W07-1106/}
}

References

21 references, 5 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Timothy Baldwin, Colin Bannard, Takaaki Tanaka, and Dominic Widdows. 2003. An empirical model of multiword expression decomposability. In Proceedings of the ACL-SIGLEX Workshop on Multiword Expressions: Analysis, Acquisition and Treatment. find our card
  2. Julia Birke and Anoop Sarkar. 2006. A clustering approach for nearly unsupervised recognition of nonliteral language. In Proceedings of EACL-06, 329–336. find
  3. Michael Collins. 1999. Head-Driven Statistical Models for Natural Language Parsing. Ph.D. thesis, University of Pennsylvania. find
  4. Anthony P. Cowie, Ronald Mackin, and Isabel R. Mc-Caig. 1983. Oxford Dictionary of Current Idiomatic English, volume 2. Oxford University Press. find
  5. Scott C. Deerwester, Susan T. Dumais, Thomas K. Landauer, George W. Furnas, and Richard A. Harshman. 1990. Indexing by latent semantic analysis. Journal of the American Society of Information Science, 41(6):391–407. find
  6. Stefan Evert, Ulrich Heid, and Kristina Spranger. 2004. Identifying morphosyntactic preferences in collocations. In Proceedings LREC-04. find
  7. Afsaneh Fazly and Suzanne Stevenson. 2006. Automatically constructing a lexicon of verb phrase idiomatic combinations. In Proceedings of EACL-06, 337–344. find our card
  8. Christiane Fellbaum. 2002. VP idioms in the lexicon: Topics for research using a very large corpus. In S. Busemann, editor, Proceedings of the KONVENS-02 Conference. find
  9. John R. Firth. 1957. A synopsis of linguistic theory 1930–1955. In Studies in Linguistic Analysis (special volume of the Philological Society), 1–32. The Philological Society, Oxford. find
  10. Chikara Hashimoto, Satoshi Sato, and Takehito Utsuro. 2006. Japanese idiom recognition: Drawing a line between literal and idiomatic meanings. In Proceedings of the COLING/ACL 2006 Main Conference Poster Sessions, 353–360. find
  11. Graham Katz and Eugenie Giesbrecht. 2006. Automatic identification of non-compositional multi-word expressions using latent semantic analysis. In Proceedings of the ACL/COLING-06 Workshop on Multiword Expressions: Identifying and Exploiting Underlying Properties, 12–19. find our card
  12. Brigitte Krenn and Stefan Evert. 2001. Can we do better than frequency? A case study on extracting PP-verb collocations. In Proceedings of the ACL-01 Workshop on Collocations. find
  13. Dekang Lin. 1999. Automatic identification of non-compositional phrases. In Proceedings of ACL-99, 317–324. find our card
  14. Diana McCarthy, Bill Keller, and John Carroll. 2003. Detecting a continuum of compositionality in phrasal verbs. In Proceedings of the ACL-SIGLEX Workshop on Multiword Expressions: Analysis, Acquisition and Treatment. find
  15. Geoffrey Nunberg, Ivan A. Sag, and Thomas Wasow. 1994. Idioms. Language, 70(3):491–538. find
  16. Jon Patrick and Jeremy Fletcher. 2005. Classifying verb-particle constructions by verb arguments. In Proceedings of the Second ACL-SIGSEM Workshop on the Linguistic Dimensions of Prepositions and their use in Computational Linguistics Formalisms and Applications, 200–209. find
  17. Ivan A. Sag, Timothy Baldwin, Francis Bond, Ann Copestake, and Dan Flickinger. 2002. Multiword expressions: A pain in the neck for NLP. In Proceedings of CICLing-02, 1–15. find our card
  18. Maggie Seaton and Alison Macaulay, editors. 2002. Collins COBUILD Idioms Dictionary. HarperCollins Publishers, second edition. find
  19. Kiyoko Uchiyama, Timothy Baldwin, and Shun Ishizaki. 2005. Disambiguating Japanese compound verbs. Computer Speech and Language, Special Issue on Multiword Expressions, 19(4):497–512. find
  20. Sriram Venkatapathy and Aravid Joshi. 2005. Measuring the relative compositionality of verb-noun (V-N) collocations by integrating features. In Proceedings of HLT/EMNLP-05, 899–906. find
  21. Begoña Villada Moirón and Jörg Tiedemann. 2006. Identifying idiomatic expressions using automatic word-alignment. In Proceedings of the EACL-06 Workshop on Multiword Expressions in a Multilingual Context, 33–40. find

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/papers/cook-2007-pulling-their-weight-exploiting-syntactic/