The Role of Modifier and Head Properties in Predicting the Compositionality of English and German Noun-Noun Compounds: A Vector-Space Perspective

Walde, Sabine Schulte im; Hätty, Anna; Bott, Stefan

doi:10.18653/v1/s16-2020

Cited by 13 publications

(16 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This might explain the observed asymmetry. This finding is in line with earlier studies (Hätty, 2016;Schulte im Walde et al, 2016a) which investigated the asymmetry between the properties of heads and modifiers in noun-noun compounds. They showed that head constituent properties, such as their ambiguity or frequency, influence the predictability of NC compositionality to a much larger degree than modifier constituent properties.…”

Section: Resultssupporting

confidence: 93%

“…As to our knowledge, few systems have attempted to distinguish between word senses and then address various semantic relatedness tasks, such as Li and Jurafsky (2015) and Iacobacci et al (2015). Computational compositionality assessment has been studied for NCs (Reddy et al, 2011;Schulte im Walde et al, 2013;Schulte im Walde et al, 2016a) and PVs (McCarthy et al, 2003;Bannard, 2005;Kühner and Schulte im Walde, 2010). Most similar to our current work is Salehi et al (2015a), who addressed the problem of semantic ambiguity in MWEs by using a multi-sense skip gram model with two to five embeddings per word.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Proceedings of the 13th Workshop on Multiword Expressions (MWE 2017)

Μαρκαντωνάτου¹,

Ramisch²,

Savary

et al. 2017

View full text Add to dashboard Cite

The workshop has been held since 2001 in conjunction with major computational linguistics conferences (ACL, COLING, LREC, EACL). It attracts the attention of an ever-growing community working on a variety of languages and linguistic phenomena and represents an important venue for the community to interact, share resources and tools, and collaborate on efforts for advancing the computational treatment of multiword expressions.In this 13th edition of the workshop, we have called for papers on major challenges in MWE processing, both from theoretical and computational viewpoints, focusing on research related (but not limited) to the following topics:• Manually and automatically constructed lexical resources • MWE representation in lexical resources • MWE annotation in corpora and treebanks • MWEs in non-standard language (e.g. tweets, forums, spontaneous speech) • Original MWE discovery methods (e.g. using word embeddings, parallel corpora)• Original MWE in-context identification methods (e.g. using deep learning, topic models) • MWE processing in syntactic frameworks (e.g. HPSG, LFG, TAG, universal dependencies) • MWE processing in semantic frameworks (e.g. WSD, semantic parsing)• MWE processing in end-user applications (e.g. summarization, machine translation) • Orchestration of MWE processing with respect to applications • Evaluation of MWE processing techniques • Models of first and second language acquisition of MWEs • Theoretical and psycholinguistic studies on MWEs • Crosslinguistic studies on MWEs Submissions included both long and short papers. In total, 34 papers (22 short and 12 long) were submitted to the main track of the workshop. From those, 7 papers (4 short and 3 long) were accepted as oral presentations and 14 papers (11 short and 3 long) were accepted as posters. The overall acceptance rate is 62% including oral presentations and posters, short and long papers.Additionally, 6 system description papers and 1 shared task description paper were submitted to the shared task track. The former were all selected as posters and the latter as an oral presentation. The reviewing modalities were different in this track, therefore we do not count these papers in the workshop acceptance rate.In addition to the oral and poster sessions, the workshop featured an invited talk by Paul Cook, from the University of New Brunswick, Canada. The presentation was entitled Exploiting multilingual lexical 1 http://multiword.sf.net/mwe2017 2 http://www.siglex.org/ 3 http://parseme.eu/ iii resources to predict the compositionality of MWEs. The program also included a panel discussion on the future directions of the MWE community and the SIGLEX Section.We would like to thank the members of the program committee for the timely reviews, authors for their valuable contributions, shared task organizers, annotators, and system developers for their hard work, and all involved participants for their interest in the workshop. We also want to thank the IC1207 COST Action PARSEME and SIGLEX for their endorsement and support, as well a...

show abstract

Section: Resultssupporting

confidence: 93%

Section: Introductionmentioning

confidence: 99%

Proceedings of the 13th Workshop on Multiword Expressions (MWE 2017)

Μαρκαντωνάτου¹,

Ramisch²,

Savary

et al. 2017

View full text Add to dashboard Cite

show abstract

“…Only 18 of the 244 compounds occurred ≥100 times, which makes the results less conclusive. For this data set, we had access to the ranking of (Schulte im Walde et al, 2016a) and thus compare our results to theirs ((mod|head).vector in Table 3). Note that the numbers given here differ from those given in (Schulte im Walde et al, 2016a) because they are not calculated on the whole VDHB dataset but only on subsets of it.…”

Section: Germanmentioning

confidence: 99%

“…Thanks to the anonymous reviewers for their constructive comments and to Schulte im Walde et al (2016a) for sharing their results to enable a direct comparison.…”

Section: Acknowledgementsmentioning

confidence: 99%

Show Me Your Variance and I Tell You Who You Are - Deriving Compound Compositionality from Word Alignments

Cap¹

2017

Proceedings of the 13th Workshop on Multiword Expressions (MWE 2017)

View full text Add to dashboard Cite

show abstract

“…As to our knowledge, few systems have attempted to distinguish between word senses and then address various semantic relatedness tasks, such as Li and Jurafsky (2015) and Iacobacci et al (2015). Computational compositionality assessment has been studied for NCs (Reddy et al, 2011;Schulte im Walde et al, 2013;Salehi and Cook, 2013;Schulte im Walde et al, 2016a) and PVs (McCarthy et al, 2003;Baldwin et al, 2003;Bannard, 2005;Kühner and Schulte im Walde, 2010). Most similar to our current work is Salehi et al (2015a), who addressed the problem of semantic ambiguity in MWEs by using a multi-sense skip gram model with two to five embeddings per word.…”

Section: Introductionmentioning

confidence: 99%

Factoring Ambiguity out of the Prediction of Compositionality for German Multi-Word Expressions

Bott¹,

Walde²

2017

Proceedings of the 13th Workshop on Multiword Expressions (MWE 2017)

Self Cite

View full text Add to dashboard Cite

Ambiguity represents an obstacle for distributional semantic models (DSMs), which typically subsume the contexts of all word senses within one vector. While individual vector space approaches have been concerned with sense discrimination (e.g., Schütze (1998), Erk (2009), Erk and Pado (2010)), such discrimination has rarely been integrated into DSMs across semantic tasks. This paper presents a softclustering approach to sense discrimination that filters sense-irrelevant features when predicting the degrees of compositionality for German noun-noun compounds and German particle verbs.

show abstract

The Role of Modifier and Head Properties in Predicting the Compositionality of English and German Noun-Noun Compounds: A Vector-Space Perspective

Cited by 13 publications

References 28 publications

Proceedings of the 13th Workshop on Multiword Expressions (MWE 2017)

Proceedings of the 13th Workshop on Multiword Expressions (MWE 2017)

Show Me Your Variance and I Tell You Who You Are - Deriving Compound Compositionality from Word Alignments

Factoring Ambiguity out of the Prediction of Compositionality for German Multi-Word Expressions

Contact Info

Product

Resources

About