Show simple item record

dc.contributor.authorWilson, Stevenen
dc.contributor.authorMagdy, Waliden
dc.contributor.authorMcGillivray, Barbaraen
dc.contributor.authorTyson, Garethen
dc.date.accessioned2021-01-08T00:32:00Z
dc.date.available2021-01-08T00:32:00Z
dc.identifier.isbn9781952148668en
dc.identifier.urihttps://www.repository.cam.ac.uk/handle/1810/315896
dc.description.abstractPrevious work has shown how to effectively use external resources such as dictionaries to improve English-language word embeddings, either by manipulating the training process or by applying post-hoc adjustments to the embedding space. We experiment with a multi-task learning approach for explicitly incorporating the structured elements of dictionary entries, such as user-assigned tags and usage examples, when learning embeddings for dictionary headwords. Our work generalizes several existing models for learning word embeddings from dictionaries. However, we find that the most effective representations overall are learned by simply training with a skip-gram objective over the concatenated text of all entries in the dictionary, giving no particular focus to the structure of the entries.
dc.publisherAssociation for Computational Linguistics
dc.rightsPublisher's own licence
dc.rights.uri
dc.titleEmbedding Structured Dictionary Entriesen
dc.typeConference Object
prism.endingPage125
prism.publicationNameProceedings of the First Workshop on Insights from Negative Results in NLPen
prism.startingPage117
dc.identifier.doi10.17863/CAM.63007
dcterms.dateAccepted2020-10-01en
rioxxterms.versionVoR
rioxxterms.licenseref.urihttp://www.rioxx.net/licenses/all-rights-reserveden
rioxxterms.licenseref.startdate2020-10-01en
dc.contributor.orcidMcGillivray, Barbara [0000-0003-3426-8200]
rioxxterms.typeConference Paper/Proceeding/Abstracten
pubs.funder-project-idAlan Turing Institute (EP/N510129/1)
dc.identifier.urlhttps://www.aclweb.org/anthology/2020.insights-1.18.pdfen
pubs.conference-nameFirst Workshop on Insights from Negative Results in NLPen
pubs.conference-start-date2020-11-19en


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record