Combined approach to hypernym detection for thesaurus enrichment

dc.contributor.author	Tikhomirov M.M.
dc.contributor.author	Loukachevitch N.V.
dc.contributor.author	Parkhomenko E.A.
dc.date.accessioned	2021-02-25T07:01:55Z
dc.date.available	2021-02-25T07:01:55Z
dc.date.issued	2020
dc.identifier.issn	2221-7932
dc.identifier.uri	https://dspace.kpfu.ru/xmlui/handle/net/161618
dc.description.abstract	© 2020 ABBYY PRODUCTION LLC. All rights reserved. This paper describes a combined approach to hypernym detection task. The approach combines the following techniques: distribution semantics, rule-based patterns, and modern neural networks (BERT). An important feature of our solution is that hypernyms are extracted only from a single text collection provided by the organizers. The described approach obtained the fourth result on the private nouns track. It was found out that the use of the rule-based patterns can significantly improve the results. Also, using the BERT model as an additional factor always helps to improve the performance.
dc.relation.ispartofseries	Komp'juternaja Lingvistika i Intellektual'nye Tehnologii
dc.subject	BERT
dc.subject	Embeddings
dc.subject	Hypernym extraction
dc.subject	Patterns
dc.subject	Thesaurus
dc.title	Combined approach to hypernym detection for thesaurus enrichment
dc.type	Conference Paper
dc.relation.ispartofseries-issue	19
dc.relation.ispartofseries-volume	2020-June
dc.collection	Публикации сотрудников КФУ
dc.relation.startpage	736
dc.source.id	SCOPUS22217932-2020-2020-19-SID85093869167

Файлы в этом документе

Имя: SCOPUS22217932-20 ...

Размер: 49.97Kb

Формат: PDF

Публикации сотрудников КФУ Scopus [24551]
Коллекция содержит публикации сотрудников Казанского федерального (до 2010 года Казанского государственного) университета, проиндексированные в БД Scopus, начиная с 1970г.