Elamur Research⌕
☰
home / papers / 2009 / Unsupervised Recognition of Literal and Non-Literal Use of Idiomatic Expressions

Unsupervised Recognition of Literal and Non-Literal Use of Idiomatic Expressions

Summary

The paper tackles token-level idiom classification: deciding whether a potentially idiomatic expression such as spill the beans or play with fire is meant literally or figuratively in a specific context. Most earlier work instead classified expression types as idioms or non-idioms. The authors adapt Hirst and St-Onge's cohesion-based idea for detecting malapropisms. When an expression's component words connect semantically to the surrounding discourse, a literal reading is likely. When they do not, the usage is probably idiomatic. Relatedness is computed without hand-built resources like WordNet, using Normalized Google Distance (NGD), which relies on search-engine page counts. Two classifiers are proposed. One builds lexical chains and needs parameter tuning. The other, fully unsupervised, checks whether removing the expression increases the connectivity of a cohesion graph. The graph-based method matches the optimally tuned chain classifier and outperforms the CForm and Rep baselines. It also handles expressions in canonical form, which the CForm approach always labels as idiomatic. This makes cohesion a practical, knowledge-lean cue for literal versus non-literal disambiguation.

Key points

  • The evaluation covers 17 English expressions, including break the ice, drop the ball, play with fire and spill the beans. Some of them, such as drop the ball, pull the trigger and bounce off the wall, are mostly used literally.
  • The graph-based classifier and the globally optimised chain classifier (LC-O) both reach a literal-class F-Score of about 60%, with recall above precision.
  • The chain classifier tuned realistically on separate development data (break the ice) scores much lower, because it seldom predicts the literal class.
  • A simple supervised classifier still beats all unsupervised methods by a wide margin.
  • Per-idiom accuracy of the graph-based method is above 50% for most expressions, peaking at 98.95% for set in stone. The lowest score is 11.11% for blow one's own trumpet, which has only 9 instances.
  • The method is not biased toward the idiomatic reading. For drop the ball, 472 of 530 literal predictions were correct.

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

69 papers of our library cite it (of the 140673 whose bibliographies we read); 11 of them have a page here, 58 not yet.

Cite

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

@inproceedings{sporleder2009unsupervised,
  title = {{Unsupervised Recognition of Literal and Non-Literal Use of Idiomatic Expressions}},
  author = {Sporleder, Caroline and Li, Linlin},
  booktitle = {Proceedings of the 12th Conference of the European Chapter of the ACL (EACL 2009)},
  year = {2009},
  doi = {10.3115/1609067.1609151},
  url = {https://doi.org/10.3115/1609067.1609151}
}

References

27 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 2003 workshop on Multiword expressions: analysis, acquisition and treatment, pages 89–96. find our card
  2. Colin Bannard. 2007. A measure of syntactic flexibility for automatically identifying multiword expressions in corpora. In Proceedings of the ACL-07 Workshop on A Broader Perspective on Multiword Expressions, pages 1–8. find
  3. Regina Barzilay and Michael Elhadad. 1997. Using lexical chains for text summarization. In Proceedings of the ACL-97 Intelligent Scalable Text Summarization Workshop (ISTS-97). find
  4. Julia Birke and Anoop Sarkar. 2006. A clustering approach for the nearly unsupervised recognition of nonliteral language. In Proceedings of EACL-06, pages 329–336. find
  5. Alexander Budanitsky and Graeme Hirst. 2006. Evaluating WordNet-based measures of semantic distance. Computational Linguistics, 32(1):13–47. find
  6. Rudi L. Cilibrasi and Paul M.B. Vitanyi. 2007. The Google similarity distance. IEEE Trans. Knowledge and Data Engineering, 19(3):370–383. find
  7. Jacob Cohen. 1960. A coefficient of agreement for nominal scales. Educational and Psychological Measurements, 20:37–46. find
  8. Paul Cook, Afsaneh Fazly, and Suzanne Stevenson. 2007. Pulling their weight: Exploiting syntactic forms for the automatic identification of idiomatic expressions in context. In Proceedings of the ACL-07 Workshop on A Broader Perspective on Multiword Expressions, pages 41–48. find our card
  9. A.P. Cowie, R. Mackin, and I.R. McCaig. 1997. Oxford dictionary of English idioms. Oxford University Press. find
  10. Afsaneh Fazly and Suzanne Stevenson. 2006. Automatically constructing a lexicon of verb phrase idiomatic combinations. In Proceedings of EACL-06. find our card
  11. Afsaneh Fazly, Paul Cook, and Suzanne Stevenson. To appear. Unsupervised type and token identification of idiomatic expressions. Computational Linguistics. find our card
  12. M.A.K. Halliday and R. Hasan. 1976. Cohesion in English. Longman House, New York. find
  13. Donald Hindle. 1990. Noun classification from predicate-argument structures. In Proceedings of ACL-90, pages 268–275. find
  14. Graeme Hirst and David St-Onge. 1998. Lexical chains as representations of context for the detection and correction of malapropisms. In Christiane Fellbaum, editor, WordNet: An electronic lexical database, pages 305–332. The MIT Press. find
  15. Graham Katz and Eugenie Giesbrecht. 2006. Automatic identification of non-compositional multiword expressions using latent semantic analysis. In Proceedings of the ACL/COLING-06 Workshop on Multiword Expressions: Identifying and Exploiting Underlying Properties, pages 12–19. find
  16. Mirella Lapata and Frank Keller. 2005. Web-based models for natural language processing. ACM Transactions on Speech and Language Processing, 2:1–31. find
  17. Dekang Lin. 1998. Automatic retrieval and clustering of similar words. In Proceedings of ACL-98, pages 768–774. find
  18. Dekang Lin. 1999. Automatic identification of non-compositional phrases. In Proceedings of ACL-99, pages 317–324. find our card
  19. Guido Minnen, John Carroll, and Darren Pearce. 2001. Applied morphological processing of English. Natural Language Engineering, 7(3):207–223. find
  20. Saif Mohammad and Graeme Hirst. 2006. Distributional measures of concept-distance: A task-oriented evaluation. In Proceedings of EMNLP-06. find
  21. Jane Morris and Graeme Hirst. 2004. Non-classical lexical semantic relations. In HLT-NAACL-04 Workshop on Computational Lexical Semantics, pages 46–51. find
  22. Simone Paolo Ponzetto and Michael Strube. 2007. Knowledge derived from Wikipedia for computing semantic relatedness. Journal of Artificial Intelligence Research, 30:181–212. find
  23. Susanne Riehemann. 2001. A Constructional Approach to Idioms and Word Formation. Ph.D. thesis, Stanford University. find
  24. H. Gregory Silber and Kathleen F. McCoy. 2002. Efficiently computed lexical chains as an intermediate representation for automatic text summarization. Computational Linguistics, 28(4):487–496. find
  25. Aline Villavicencio, Valia Kordoni, Yi Zhang, Marco Idiart, and Carlos Ramisch. 2007. Validation and evaluation of automatically acquired multiword expressions for grammar engineering. In Proceedings of EMNLP-07, pages 1034–1043. find
  26. Torsten Zesch, Christof Müller, and Iryna Gurevych. 2008. Using wiktionary for computing semantic relatedness. In Proceedings of AAAI-08, pages 861– 867. find
  27. Xiaojin Zhu and Ronald Rosenfeld. 2001. Improving trigram language modeling with the world wide web. In Proceedings of ICASSP-01. find

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/papers/sporleder-2009-unsupervised-recognition-literal-non-literal/