Cargando…

A transition-based neural framework for Chinese information extraction

Chinese information extraction is traditionally performed in the process of word segmentation, entity recognition, relation extraction and event detection. This pipelined approach suffers from two limitations: 1) It is prone to introduce propagated errors from upstream tasks to subsequent applicatio...

Descripción completa

Detalles Bibliográficos
Autores principales: Huang, Wenzhi, Zhang, Junchi, Ji, Donghong
Formato: Online Artículo Texto
Lenguaje:English
Publicado: Public Library of Science 2020
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7363078/
https://www.ncbi.nlm.nih.gov/pubmed/32667950
http://dx.doi.org/10.1371/journal.pone.0235796
_version_ 1783559603497205760
author Huang, Wenzhi
Zhang, Junchi
Ji, Donghong
author_facet Huang, Wenzhi
Zhang, Junchi
Ji, Donghong
author_sort Huang, Wenzhi
collection PubMed
description Chinese information extraction is traditionally performed in the process of word segmentation, entity recognition, relation extraction and event detection. This pipelined approach suffers from two limitations: 1) It is prone to introduce propagated errors from upstream tasks to subsequent applications; 2) Mutual benefits of cross-task dependencies are hard to be introduced in non-overlapping models. To address these two challenges, we propose a novel transition-based model that jointly performs entity recognition, relation extraction and event detection as a single task. In addition, we incorporate subword-level information into character sequence with the use of a hybrid lattice structure, removing the reliance of external word tokenizers. Results on standard ACE benchmarks show the benefits of the proposed joint model and lattice network, which gives the best result in the literature.
format Online
Article
Text
id pubmed-7363078
institution National Center for Biotechnology Information
language English
publishDate 2020
publisher Public Library of Science
record_format MEDLINE/PubMed
spelling pubmed-73630782020-07-23 A transition-based neural framework for Chinese information extraction Huang, Wenzhi Zhang, Junchi Ji, Donghong PLoS One Research Article Chinese information extraction is traditionally performed in the process of word segmentation, entity recognition, relation extraction and event detection. This pipelined approach suffers from two limitations: 1) It is prone to introduce propagated errors from upstream tasks to subsequent applications; 2) Mutual benefits of cross-task dependencies are hard to be introduced in non-overlapping models. To address these two challenges, we propose a novel transition-based model that jointly performs entity recognition, relation extraction and event detection as a single task. In addition, we incorporate subword-level information into character sequence with the use of a hybrid lattice structure, removing the reliance of external word tokenizers. Results on standard ACE benchmarks show the benefits of the proposed joint model and lattice network, which gives the best result in the literature. Public Library of Science 2020-07-15 /pmc/articles/PMC7363078/ /pubmed/32667950 http://dx.doi.org/10.1371/journal.pone.0235796 Text en © 2020 Huang et al http://creativecommons.org/licenses/by/4.0/ This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0/) , which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
spellingShingle Research Article
Huang, Wenzhi
Zhang, Junchi
Ji, Donghong
A transition-based neural framework for Chinese information extraction
title A transition-based neural framework for Chinese information extraction
title_full A transition-based neural framework for Chinese information extraction
title_fullStr A transition-based neural framework for Chinese information extraction
title_full_unstemmed A transition-based neural framework for Chinese information extraction
title_short A transition-based neural framework for Chinese information extraction
title_sort transition-based neural framework for chinese information extraction
topic Research Article
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7363078/
https://www.ncbi.nlm.nih.gov/pubmed/32667950
http://dx.doi.org/10.1371/journal.pone.0235796
work_keys_str_mv AT huangwenzhi atransitionbasedneuralframeworkforchineseinformationextraction
AT zhangjunchi atransitionbasedneuralframeworkforchineseinformationextraction
AT jidonghong atransitionbasedneuralframeworkforchineseinformationextraction
AT huangwenzhi transitionbasedneuralframeworkforchineseinformationextraction
AT zhangjunchi transitionbasedneuralframeworkforchineseinformationextraction
AT jidonghong transitionbasedneuralframeworkforchineseinformationextraction