Cargando…

Position-Enhanced Multi-Head Self-Attention Based Bidirectional Gated Recurrent Unit for Aspect-Level Sentiment Classification

Aspect-level sentiment classification (ASC) is an interesting and challenging research task to identify the sentiment polarities of aspect words in sentences. Previous attention-based methods rarely consider the position information of aspect and contextual words. For an aspect word in a sentence, i...

Descripción completa

Detalles Bibliográficos
Autores principales: Li, Xianyong, Ding, Li, Du, Yajun, Fan, Yongquan, Shen, Fashan
Formato: Online Artículo Texto
Lenguaje:English
Publicado: Frontiers Media S.A. 2022
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8822055/
https://www.ncbi.nlm.nih.gov/pubmed/35145460
http://dx.doi.org/10.3389/fpsyg.2021.799926
_version_ 1784646529961689088
author Li, Xianyong
Ding, Li
Du, Yajun
Fan, Yongquan
Shen, Fashan
author_facet Li, Xianyong
Ding, Li
Du, Yajun
Fan, Yongquan
Shen, Fashan
author_sort Li, Xianyong
collection PubMed
description Aspect-level sentiment classification (ASC) is an interesting and challenging research task to identify the sentiment polarities of aspect words in sentences. Previous attention-based methods rarely consider the position information of aspect and contextual words. For an aspect word in a sentence, its adjacent words should be given more attention than the long distant words. Based on this consideration, this article designs a position influence vector to represent the position information between an aspect word and the context. By combining the position influence vector, multi-head self-attention mechanism and bidirectional gated recurrent unit (BiGRU), a position-enhanced multi-head self-attention network based BiGRU (PMHSAT-BiGRU) model is proposed. To verify the effectiveness of the proposed model, this article makes a large number of experiments on SemEval2014 restaurant, SemEval2014 laptop, SemEval2015 restaurant, and SemEval2016 restaurant data sets. The experiment results show that the performance of the proposed PMHSAT-BiGRU model is obviously better than the baselines. Specially, compared with the original LSTM model, the Accuracy values of the proposed PMHSAT-BiGRU model on the four data sets are improved by 5.72, 6.06, 4.52, and 3.15%, respectively.
format Online
Article
Text
id pubmed-8822055
institution National Center for Biotechnology Information
language English
publishDate 2022
publisher Frontiers Media S.A.
record_format MEDLINE/PubMed
spelling pubmed-88220552022-02-09 Position-Enhanced Multi-Head Self-Attention Based Bidirectional Gated Recurrent Unit for Aspect-Level Sentiment Classification Li, Xianyong Ding, Li Du, Yajun Fan, Yongquan Shen, Fashan Front Psychol Psychology Aspect-level sentiment classification (ASC) is an interesting and challenging research task to identify the sentiment polarities of aspect words in sentences. Previous attention-based methods rarely consider the position information of aspect and contextual words. For an aspect word in a sentence, its adjacent words should be given more attention than the long distant words. Based on this consideration, this article designs a position influence vector to represent the position information between an aspect word and the context. By combining the position influence vector, multi-head self-attention mechanism and bidirectional gated recurrent unit (BiGRU), a position-enhanced multi-head self-attention network based BiGRU (PMHSAT-BiGRU) model is proposed. To verify the effectiveness of the proposed model, this article makes a large number of experiments on SemEval2014 restaurant, SemEval2014 laptop, SemEval2015 restaurant, and SemEval2016 restaurant data sets. The experiment results show that the performance of the proposed PMHSAT-BiGRU model is obviously better than the baselines. Specially, compared with the original LSTM model, the Accuracy values of the proposed PMHSAT-BiGRU model on the four data sets are improved by 5.72, 6.06, 4.52, and 3.15%, respectively. Frontiers Media S.A. 2022-01-25 /pmc/articles/PMC8822055/ /pubmed/35145460 http://dx.doi.org/10.3389/fpsyg.2021.799926 Text en Copyright © 2022 Li, Ding, Du, Fan and Shen. https://creativecommons.org/licenses/by/4.0/This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
spellingShingle Psychology
Li, Xianyong
Ding, Li
Du, Yajun
Fan, Yongquan
Shen, Fashan
Position-Enhanced Multi-Head Self-Attention Based Bidirectional Gated Recurrent Unit for Aspect-Level Sentiment Classification
title Position-Enhanced Multi-Head Self-Attention Based Bidirectional Gated Recurrent Unit for Aspect-Level Sentiment Classification
title_full Position-Enhanced Multi-Head Self-Attention Based Bidirectional Gated Recurrent Unit for Aspect-Level Sentiment Classification
title_fullStr Position-Enhanced Multi-Head Self-Attention Based Bidirectional Gated Recurrent Unit for Aspect-Level Sentiment Classification
title_full_unstemmed Position-Enhanced Multi-Head Self-Attention Based Bidirectional Gated Recurrent Unit for Aspect-Level Sentiment Classification
title_short Position-Enhanced Multi-Head Self-Attention Based Bidirectional Gated Recurrent Unit for Aspect-Level Sentiment Classification
title_sort position-enhanced multi-head self-attention based bidirectional gated recurrent unit for aspect-level sentiment classification
topic Psychology
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8822055/
https://www.ncbi.nlm.nih.gov/pubmed/35145460
http://dx.doi.org/10.3389/fpsyg.2021.799926
work_keys_str_mv AT lixianyong positionenhancedmultiheadselfattentionbasedbidirectionalgatedrecurrentunitforaspectlevelsentimentclassification
AT dingli positionenhancedmultiheadselfattentionbasedbidirectionalgatedrecurrentunitforaspectlevelsentimentclassification
AT duyajun positionenhancedmultiheadselfattentionbasedbidirectionalgatedrecurrentunitforaspectlevelsentimentclassification
AT fanyongquan positionenhancedmultiheadselfattentionbasedbidirectionalgatedrecurrentunitforaspectlevelsentimentclassification
AT shenfashan positionenhancedmultiheadselfattentionbasedbidirectionalgatedrecurrentunitforaspectlevelsentimentclassification