English Multiword Expression-aware Dependency Parsing Including Named Entities

What it is about

In Penn-style treebanks, multiword expressions (MWEs) and named entities are annotated independently of the syntax, so their spans often cross the phrase structure, and dependency trees converted from them disagree with the MWE spans. The authors build a corpus from OntoNotes in which functional MWEs (prepositions, conjunctions, adverbs and so on) and named entities are subtrees, and convert it to dependencies in which MWE-internal links carry MWE-specific labels. They then compare a pipeline (a CRF finds MWE spans, then a parser) with a joint model that predicts spans and an MWE-aware tree in one step with a transition parser (Redshift, beam 16). Adding MWE-span features to the joint model gives a 1.35-point gain in MWE recognition (FUM) over the pipeline; the labelled attachment of the first tokens of MWEs improves by 3.0 points over the joint model without them. The LAS on all sentences is about 89.5 in both systems.

Ideas

• I1. en/seeds/lexicon and en/src/tok.rs: merge named-entity and fixed MWE spans into one unit before parsing, with flat and fixed relations as in the converted corpus; metric: LAS on EWT test and ATIS (current LAS 63.0), with the MWE-internal attachments scored separately.
• I2. en/src/parse.rs: add MWE-specific labels (fixed, flat, compound:prt-style) and span features to the arc-standard perceptron, so that span and tree are predicted together; metric: MWE span F on EWT test and the first-token LAS, two runs for determinism.
• I3. Dictionary matching against a predicted span: the paper finds span prediction beats dictionary matching by 0.8 points; compare en/seeds/lexicon matching with a perceptron span predictor on EWT test before adding the joint model.

For Mova

The chapter is 19. The lexicon: vocabulary, multiword expressions and register. The code is en/src/tok.rs (unit merging), en/src/parse.rs (joint labels) and the seed lexicon in en/seeds/lexicon. The Wiktionary-derived functional MWE list in the paper is a licence question: take the method, not the data, unless the source licence allows it.

Topics: Named entity recognition · Dependency parsing

Cite

The paper itself, not this card. BibTeX from the library's record of the work. DOI 10.18653/v1/p17-2068

@inproceedings{kato2017english,
  title = {{English Multiword Expression-aware Dependency Parsing Including Named Entities}},
  author = {Kato, Akihiko and Shindo, Hiroyuki and Matsumoto, Yuji},
  booktitle = {Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers)},
  year = {2017},
  doi = {10.18653/v1/p17-2068},
  url = {https://doi.org/10.18653/v1/p17-2068}
}

References

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

1. Anne Abeillé, Lionel Clément, and François Toussenel. 2003. Building a Treebank for French, Springer Netherlands, Dordrecht, pages 165–187. find
2. Korkontzelos and Manandhar (2010) reports an improvement in base-phrase chunking by pre-grouping MWEs as words-with-spaces. They focus on compound nouns, adjective-noun constructions, and named entities. However, they use gold MWE-spans, and this is not a realistic setting. By contrast, we use predicted MWE-spans. find
3. Ann Bies, Justin Mott, Colin Warner, and Seth Kulick. 2012. English web treebank. Technical Report LDC2012T13, Linguistic Data Consortium, Philadelphia, Pennsylvania, USA. . find
4. Marie Candito and Matthieu Constant. 2014. Strategies for contiguous multiword expression analysis and dependency parsing. In Proceedings of the 52nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). Association for Computational Linguistics, pages 743–753. https://doi.org/10.3115/v1/P14-1070. Three works concerned with a French MWE-aware syntactic parsing are relevant. ↗
5. First, Green et al. (2013) proposes a method for recognizing contiguous MWEs as a part of constituency parsing by using MWE-specific non-terminals. find
6. They Rose Jenny Finkel and D. Christopher Manning. 2009. Joint parsing and named entity recognition. In Proceedings of Human Language Technologies: The 2009 Annual Conference of the North American Chapter of the Association for Computational Linguistics. Association for Computational Linguistics, pages 326–334. http://aclweb.org/anthology/N09-1037. ↗
7. Bertomeu Castelló, and Jungmee Lee. 2013. Universal dependency annotation for multilingual parsing. In Proceedings of the 51st Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers). Association for Computational Linguistics, pages 92–97. http://aclweb.org/anthology/P13-2017. ↗
8. Alexis Nasr, Carlos Ramisch, José Deulofeu, and André Valli. 2015. Joint dependency parsing and multiword expression tokenization. In Proceedings of the 53rd Annual Meeting of the Association for Computational Linguistics and the 7th International Joint Conference on Natural Language Processing (Volume 1: Long Papers). Association for Computational Linguistics, Beijing, China, pages 1116– 1126. http://www.aclweb.org/anthology/P15-1108. ↗
9. Spence Green, Marie-Catherine de Marneffe, and Christopher D Manning. 2013. Parsing models for identifying multiword expressions. Computational Linguistics 39(1):195–227. find
10. Matthew Honnibal, Yoav Goldberg, and Mark Johnson. 2013. Proceedings of the Seventeenth Conference on Computational Natural Language Learning, Association for Computational Linguistics, chapter A Non-Monotonic Arc-Eager Transition System for Dependency Parsing, pages 163–172. http://aclweb.org/anthology/W13-3518. ↗ our card
11. Sameer S. Pradhan, Eduard H. Hovy, Mitchell P. Marcus, Martha Palmer, Lance A. Ramshaw, and Ralph M. Weischedel. 2007. Ontonotes: A unified relational semantic representation. In Proceedings of the First IEEE International Conference on Semantic Computing (ICSC 2007), September 17-19, 2007, Irvine, California, USA. pages 517–526. https://doi.org/10.1109/ICSC.2007.83. ↗
12. Akihiko Kato, Hiroyuki Shindo, and Yuji Matsumoto. 2016. Construction of an english dependency corpus incorporating compound function words. In Nicoletta Calzolari (Conference Chair), Khalid Choukri, Thierry Declerck, Sara Goggi, Marko Grobelnik, Bente Maegaard, Joseph Mariani, Helene Mazo, Asuncion Moreno, Jan Odijk, and Stelios Piperidis, editors, Proceedings of the Tenth International Conference on Language Resources and Evaluation (LREC 2016). European Language Resources Association (ELRA), Paris, France. find our card
13. Nathan Schneider, Spencer Onuffer, Nora Kazour, Emily Danchik, Michael T. Mordowanec, Henrietta Conrad, and Noah A. Smith. 2014. Comprehensive annotation of multiword expressions in a social web corpus. In Nicoletta Calzolari, Khalid Choukri, Thierry Declerck, Hrafn Loftsson, Bente Maegaard, Joseph Mariani, Asuncion Moreno, Jan Odijk, and Stelios Piperidis, editors, Proceedings of the Ninth International Conference on Language Resources and Evaluation (LREC’14). European Language Resources Association (ELRA), Reykjavik, Iceland, pages 455–461. ACL Anthology Identifier: L14-1433. ↗ our card
14. Ioannis Korkontzelos and Suresh Manandhar. 2010. Can recognising multiword expressions improve shallow parsing? In Human Language Technologies: The 2010 Annual Conference of the North American Chapter of the Association for Computational Linguistics. Association for Computational Linguistics, Los Angeles, California, pages 636– 644. http://www.aclweb.org/anthology/N10-1089. ↗
15. Yutaro Shigeto, Ai Azuma, Sorami Hisamoto, Shuhei Kondo, Tomoya Kouse, Keisuke Sakaguchi, Akifumi Yoshimoto, Frances Yung, and Yuji Matsumoto. 2013. Proceedings of the 9th Workshop on Multiword Expressions, Association for Computational Linguistics, chapter Construction of English MWE Dictionary and its Application to POS Tagging, pages 139–144. http://aclweb.org/anthology/W13-1021. ↗ our card
16. John Lafferty, Andrew McCallum, and Fernando Pereira. 2001. Conditional random fields: Probabilistic models for segmenting and labeling sequence data. In Proc. 18th International Conf. on Machine Learning. Morgan Kaufmann, San Francisco, CA, pages 282–289. citeseer.ist.psu.edu/lafferty01conditional.html. find
17. Kristina Toutanova, Dan Klein, Christopher D. Manning, and Yoram Singer. 2003. Feature-rich part-of-speech tagging with a cyclic dependency network. In Proceedings of the 2003 Human Language Technology Conference of the North American Chapter of the Association for Computational Linguistics. http://aclweb.org/anthology/N03-1033. ↗
18. Marie-Catherine Marneffe and D. Christopher Manning. 2008. Coling 2008: Proceedings of the workshop on Cross-Framework and Cross-Domain Parser Evaluation, Coling 2008 Organizing Committee, chapter The Stanford Typed Dependencies Representation, pages 1–8. http://aclweb.org/anthology/W08-1301. ↗ our card
19. Yue Zhang and Joakim Nivre. 2011. Transition-based dependency parsing with rich non-local features. In Proceedings of the 49th Annual Meeting of the Association for Computational Linguistics: Human Language Technologies. Association for Computational Linguistics, pages 188–193. http://aclweb.org/anthology/P11-2033. Ryan McDonald, Joakim Nivre, Yvonne Quirmbach-Brundage, Yoav Goldberg, Dipanjan Das, Kuzman Ganchev, Keith Hall, Slav Petrov, Hao Zhang, Oscar Täckström, Claudia Bedini, Núria ↗

This page in Markdown → https://md.mova.elamur.ai/free/en/docs/papers/aclbulk-2017-kato-english-multiword-expression-aware-dependency/