Sentiment Analysis Using Multi-Head Attention Capsules With Multi-Channel CNN and Bidirectional GRU
Existing text sentiment analysis methods mostly rely on a large number of language knowledge and sentiment resources. This paper proposes the Multi-channel convolution and bidirectional GRU multi-head attention capsule (AT-MC-BiGRU-Capsule), which uses vector neurons to replace scalar neurons to mod...
Main Authors: | , , , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
IEEE
2021-01-01
|
Series: | IEEE Access |
Subjects: | |
Online Access: | https://ieeexplore.ieee.org/document/9406579/ |
_version_ | 1811273949891264512 |
---|---|
author | Yan Cheng Huan Sun Haomai Chen Meng Li Yingying Cai Zhuang Cai Jing Huang |
author_facet | Yan Cheng Huan Sun Haomai Chen Meng Li Yingying Cai Zhuang Cai Jing Huang |
author_sort | Yan Cheng |
collection | DOAJ |
description | Existing text sentiment analysis methods mostly rely on a large number of language knowledge and sentiment resources. This paper proposes the Multi-channel convolution and bidirectional GRU multi-head attention capsule (AT-MC-BiGRU-Capsule), which uses vector neurons to replace scalar neurons to model text emotions, and uses capsules to characterize text emotions. In addition, traditional methods cannot extract the multi-level features of text sequence well. Multi-head attention can encode the dependencies between words, capture sentiment words in text, and using Convolutional Neural Network (CNN) and Bidirectional gated recurrent unit network (Bi-GRU) to extract local features and global semantic features of text respectively, the global average pooling layer is introduced to obtain the multi-level feature representation of the text sequence more comprehensively. This paper selects three English datasets and one Chinese dataset in the general corpus of sentiment classification to conduct experiments, and achieves better results than other baseline models. |
first_indexed | 2024-04-12T23:08:56Z |
format | Article |
id | doaj.art-0817a6afd47c44b4a359ddb66d13fde5 |
institution | Directory Open Access Journal |
issn | 2169-3536 |
language | English |
last_indexed | 2024-04-12T23:08:56Z |
publishDate | 2021-01-01 |
publisher | IEEE |
record_format | Article |
series | IEEE Access |
spelling | doaj.art-0817a6afd47c44b4a359ddb66d13fde52022-12-22T03:12:50ZengIEEEIEEE Access2169-35362021-01-019603836039510.1109/ACCESS.2021.30739889406579Sentiment Analysis Using Multi-Head Attention Capsules With Multi-Channel CNN and Bidirectional GRUYan Cheng0https://orcid.org/0000-0002-0160-7213Huan Sun1Haomai Chen2Meng Li3Yingying Cai4Zhuang Cai5Jing Huang6School of Computer Information Engineering, Jiangxi Normal University, Nanchang, ChinaSchool of Computer Information Engineering, Jiangxi Normal University, Nanchang, ChinaSchool of Mathematics and Compute, Yuzhang Normal University, Nanchang, ChinaSchool of Computer Information Engineering, Jiangxi Normal University, Nanchang, ChinaSchool of Computer Information Engineering, Jiangxi Normal University, Nanchang, ChinaSchool of Computer Information Engineering, Jiangxi Normal University, Nanchang, ChinaSchool of Computer Information Engineering, Jiangxi Normal University, Nanchang, ChinaExisting text sentiment analysis methods mostly rely on a large number of language knowledge and sentiment resources. This paper proposes the Multi-channel convolution and bidirectional GRU multi-head attention capsule (AT-MC-BiGRU-Capsule), which uses vector neurons to replace scalar neurons to model text emotions, and uses capsules to characterize text emotions. In addition, traditional methods cannot extract the multi-level features of text sequence well. Multi-head attention can encode the dependencies between words, capture sentiment words in text, and using Convolutional Neural Network (CNN) and Bidirectional gated recurrent unit network (Bi-GRU) to extract local features and global semantic features of text respectively, the global average pooling layer is introduced to obtain the multi-level feature representation of the text sequence more comprehensively. This paper selects three English datasets and one Chinese dataset in the general corpus of sentiment classification to conduct experiments, and achieves better results than other baseline models.https://ieeexplore.ieee.org/document/9406579/Text sentiment analysismulti-head attentionconvolutional neural networkbidirectional gated recurrent unit networkssentiment capsule |
spellingShingle | Yan Cheng Huan Sun Haomai Chen Meng Li Yingying Cai Zhuang Cai Jing Huang Sentiment Analysis Using Multi-Head Attention Capsules With Multi-Channel CNN and Bidirectional GRU IEEE Access Text sentiment analysis multi-head attention convolutional neural network bidirectional gated recurrent unit networks sentiment capsule |
title | Sentiment Analysis Using Multi-Head Attention Capsules With Multi-Channel CNN and Bidirectional GRU |
title_full | Sentiment Analysis Using Multi-Head Attention Capsules With Multi-Channel CNN and Bidirectional GRU |
title_fullStr | Sentiment Analysis Using Multi-Head Attention Capsules With Multi-Channel CNN and Bidirectional GRU |
title_full_unstemmed | Sentiment Analysis Using Multi-Head Attention Capsules With Multi-Channel CNN and Bidirectional GRU |
title_short | Sentiment Analysis Using Multi-Head Attention Capsules With Multi-Channel CNN and Bidirectional GRU |
title_sort | sentiment analysis using multi head attention capsules with multi channel cnn and bidirectional gru |
topic | Text sentiment analysis multi-head attention convolutional neural network bidirectional gated recurrent unit networks sentiment capsule |
url | https://ieeexplore.ieee.org/document/9406579/ |
work_keys_str_mv | AT yancheng sentimentanalysisusingmultiheadattentioncapsuleswithmultichannelcnnandbidirectionalgru AT huansun sentimentanalysisusingmultiheadattentioncapsuleswithmultichannelcnnandbidirectionalgru AT haomaichen sentimentanalysisusingmultiheadattentioncapsuleswithmultichannelcnnandbidirectionalgru AT mengli sentimentanalysisusingmultiheadattentioncapsuleswithmultichannelcnnandbidirectionalgru AT yingyingcai sentimentanalysisusingmultiheadattentioncapsuleswithmultichannelcnnandbidirectionalgru AT zhuangcai sentimentanalysisusingmultiheadattentioncapsuleswithmultichannelcnnandbidirectionalgru AT jinghuang sentimentanalysisusingmultiheadattentioncapsuleswithmultichannelcnnandbidirectionalgru |