Multiword expressions: linguistic precision and reusability

What it is about

The paper describes how the LinGO English Resource Grammar (ERG), a broad bidirectional HPSG grammar, encodes multiword expressions (MWEs) in a database. An MWE is listed only when the ordinary grammar rules and simplex entries fail on corpus data, so the grammar does not overgenerate. Verb-particle entries are split into a compositional case (tidy up) and an idiosyncratic case (look up), with a lexical redundancy rule that the database validates. Compound nouns are mostly left compositional, which causes ambiguity with the productive noun-noun rule. Decomposable idioms are hard to store as entries that must be kept in step with their words. The paper gives no accuracy figures; its contribution is the database design and the reuse argument.

Ideas

• I1. In en/src/lexgen.rs, store each verb-particle pair as one row with a particle slot and a flag for a compositional or an idiomatic reading. Measure: en parse-eval on EWT test, LAS on the sentences that contain a particle, with and without the rows.
• I2. Mark established compounds (car park, Easter Sunday) so that the noun-noun rule does not fire on them. Measure: the compound arc accuracy on EWT test, and the count of compound readings per noun pair before and after.
• I3. Keep non-decomposable idioms as words with spaces in prag expressions, and give decomposable ones their own rule. Measure: how many listed idioms the agreer rejects on a hand-checked set of frames from world.

For Mova

The chapter is 19. The lexicon: vocabulary, multiword expressions and register, with 14. Irregular verbs for the particle verbs and 18. Compounding, conversion and the minor processes for compounds. The code is en/src/lexgen.rs and en/src/dict.rs. The paper is NC: use it as a design source; any figure must come from EWT test.

Cite

The paper itself, not this card. BibTeX from the library's record of the work. DOI 10.63317/323zr2e2bw6g

@inproceedings{copestake2002multiword,
  title = {{Multiword expressions: linguistic precision and reusability}},
  author = {Copestake, Ann and Lambeau, Fabre and Villavicencio, Aline and Bond, Francis and Baldwin, Timothy and Sag, Ivan A. and Flickinger, Dan},
  booktitle = {Proceedings of the Third International Conference on Language Resources and Evaluation (LREC'02)},
  year = {2002},
  doi = {10.63317/323zr2e2bw6g},
  url = {https://doi.org/10.63317/323zr2e2bw6g}
}

References

6 references, 0 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

1. Copestake, Ann, 2002. Implementing typed feature structure grammars. CSLI Publications, Stanford. find
2. Copestake, Ann and Alex Lascarides, 1997. Integrating symbolic and statistical representations: The lexicon-pragmatics interface. In Proceedings of the Thirty-Fifth Annual Meeting of the Association for Computational Linguistics (ACL-97). Madrid. find
3. Fellbaum, Christiane, 1998. Towards a representation of idioms in WordNet. In Proceedings of the workshop on the Use of WordNet in Natural Language Processing Systems (Coling-ACL 1998). Montreal. find
4. Nunberg, Geoffrey, Ivan A. Sag, and Thomas Wasow, 1994. Idioms. Language, 70:491–538. find
5. Riehemann, Susanne, 2001. A Constructional Approach to Idioms and Word Formation. Ph.D. thesis, Stanford. find
6. Villavicencio, Aline and Ann Copestake, 2002. Phrasal verbs and the LinGO-ERG. LinGO Working Paper No. 2002-01. find

This page in Markdown → https://md.mova.elamur.ai/free/en/docs/papers/lingbulk-2002-copestake-multiword-expressions-linguistic-precision-reusability/