Multiword Expressions: A Pain in the Neck for NLP

By Ivan A. Sag, Timothy Baldwin, Francis Bond, Ann A. Copestake, Dan Flickinger. Conference on Intelligent Text Processing and Computational Linguistics, 2002

1,322 citations recorded by the source corpus.

Subfields: Artificial Intelligence

View Paper: https://www.semanticscholar.org/paper/5504f4ec7b7b7b074e9dc557beeaf68c76b65540

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

Citations in the corpus

Sign-Based Construction Grammar: An Informal Synopsis (2012, background): “Sag et al. (2003) introduce this category as a way of incorporating a number of proposals that have been made (e.”
From key words to key semantic domains (2008, background): “Fourth, multiword expressions (Sag et al. 2002) are not counted together.”
Word Cloud Explorer: Text Analytics Based on Word Clouds (2014, methodology): “With this simple heuristic, we can detect most compound nominals and proper names in the text (see [23] for a comprehensive summary of different multiword phenomena).”
Universal Dependencies v2: An Evergrowing Multilingual Treebank Collection (2020, background): “The name and rough scope of usage is borrowed from the ﬁxed expressions category of Sag et al. (2002).”
Universal Dependencies (2021, background): “The name and rough scope of usage is borrowed from the fixed expressions category of Sag et al. (2002).”
A Natural Proof System for Natural Language (2017, background): “The works that describe these challenges include (Sag et al., 2001), (Baldwin et al.”
Multiword Expressions (2010, background): “Source(s): Kastovsky [1982], Jackendoff [1997], Sag et al. [2002]”
Natural language based financial forecasting: a survey (2017, background): “Other popular open domain sources include SenticNet [18], which contains entries at the concept level to tackle the problem of phrases and multiword expressions [99,15].”
Towards a Seamless Integration of Word Senses into Downstream NLP Applications (2017, background): “MWE are lexical units made up of two or more words which are idiosyncratic in nature (Sag et al., 2002), e.g, Lewis Hamilton , Nico Rosberg and Formula 1 .”
Low Resource Multi-Task Sequence Tagging - Revisiting Dynamic Conditional Random Fields (2020, background): “MWEs consist of single- and multi-word noun and verb expressions with supersenses that encompass idioms, light verb constructions, verbparticle constructions, and compounds (Sag et al., 2002).”
Measuring Corporate Culture Using Machine Learning (2020, methodology): “…in the Universal Dependencies annotations, a collaborative project that defines the syntactic structure for all human languages (Nivre et al. 2016): ○ Multi-word expression (MWE): MWEs are fixed expressions that behave like function words (Sag et al. 2002).”
SemEval-2022 Task 2: Multilingual Idiomaticity Detection and Sentence Embedding (2022, background): “One distinctive aspect of MWEs is that they fall on a continuum of idiomaticity (Sag et al., 2002; Fazly et al., 2009; King and Cook, 2017), as their meaning may or may not be inferred from one of their constituents (e.”
South-East European Times : A parallel corpus of Balkan languages , Francis Tyers and (2010, background): “Indeed, Jackendoff (1997) estimates that MWEs correspond roughly to half of the entries in the lexicon of a native speaker, but the proportion of MWEs in natural language may be even larger in specialized domains or sub-languages (Sag et al., 2002).”