A Sentiment Analysis Method of Capsule Network Based on BiLSTM
Nowadays, capsule network model is widely used in image processing, whose feature engineering is not suitable for sentiment analysis based on texts obviously. In this paper, we propose a capsule network model with BiLSTM named caps-BiLSTM for sentiment analysis to solve the problem, and introduce th...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2020-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/9007445/ |
| _version_ | 1818433519813132288 |
|---|---|
| author | Yongfeng Dong Yu Fu Liqin Wang Yunliang Chen Yao Dong Jianxin Li |
| author_facet | Yongfeng Dong Yu Fu Liqin Wang Yunliang Chen Yao Dong Jianxin Li |
| author_sort | Yongfeng Dong |
| collection | DOAJ |
| description | Nowadays, capsule network model is widely used in image processing, whose feature engineering is not suitable for sentiment analysis based on texts obviously. In this paper, we propose a capsule network model with BiLSTM named caps-BiLSTM for sentiment analysis to solve the problem, and introduce the experimental results on different datasets. At the beginning of caps-BiLSTM, a convolution layer is used to transform the instance to hide vector. Then the capsule module constructs the capsule representation to the n-gram model. The state probability of a certain capsule is calculated by the capsule model. If the state probability of a given instance is the largest among all capsules, a higher coupling coefficient is assigned. Finally, in order to fusion the data features, the output of the capsule enters into a BiLSTM structure, which is used as a decoder to get the probability representation. Experimental results based on MR, IMDB and SST datasets show that the proposed method can achieve better performances than the traditional machine learning methods and the compared deeping learning models. |
| first_indexed | 2024-12-14T16:22:23Z |
| format | Article |
| id | doaj.art-901c6798c4f0498d8274154ca35b3f42 |
| institution | Directory Open Access Journal |
| issn | 2169-3536 |
| language | English |
| last_indexed | 2024-12-14T16:22:23Z |
| publishDate | 2020-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj.art-901c6798c4f0498d8274154ca35b3f422022-12-21T22:54:46ZengIEEEIEEE Access2169-35362020-01-018370143702010.1109/ACCESS.2020.29737119007445A Sentiment Analysis Method of Capsule Network Based on BiLSTMYongfeng Dong0Yu Fu1https://orcid.org/0000-0003-2335-535XLiqin Wang2Yunliang Chen3Yao Dong4Jianxin Li5https://orcid.org/0000-0002-9059-330XSchool of Artificial Intelligence, Hebei University of Technology, Tianjin, ChinaSchool of Artificial Intelligence, Hebei University of Technology, Tianjin, ChinaSchool of Artificial Intelligence, Hebei University of Technology, Tianjin, ChinaSchool of Computer Science, China University of Geosciences, Hubei, ChinaSchool of Artificial Intelligence, Hebei University of Technology, Tianjin, ChinaSmart Networks Data Laboratory, Faculty of Science, Engineering and Built Environment, School of Information Technology, Deakin University, Burwood, VIC, AustraliaNowadays, capsule network model is widely used in image processing, whose feature engineering is not suitable for sentiment analysis based on texts obviously. In this paper, we propose a capsule network model with BiLSTM named caps-BiLSTM for sentiment analysis to solve the problem, and introduce the experimental results on different datasets. At the beginning of caps-BiLSTM, a convolution layer is used to transform the instance to hide vector. Then the capsule module constructs the capsule representation to the n-gram model. The state probability of a certain capsule is calculated by the capsule model. If the state probability of a given instance is the largest among all capsules, a higher coupling coefficient is assigned. Finally, in order to fusion the data features, the output of the capsule enters into a BiLSTM structure, which is used as a decoder to get the probability representation. Experimental results based on MR, IMDB and SST datasets show that the proposed method can achieve better performances than the traditional machine learning methods and the compared deeping learning models.https://ieeexplore.ieee.org/document/9007445/Sentiment analysisneural networkcapsule networkBiLSTM |
| spellingShingle | Yongfeng Dong Yu Fu Liqin Wang Yunliang Chen Yao Dong Jianxin Li A Sentiment Analysis Method of Capsule Network Based on BiLSTM IEEE Access Sentiment analysis neural network capsule network BiLSTM |
| title | A Sentiment Analysis Method of Capsule Network Based on BiLSTM |
| title_full | A Sentiment Analysis Method of Capsule Network Based on BiLSTM |
| title_fullStr | A Sentiment Analysis Method of Capsule Network Based on BiLSTM |
| title_full_unstemmed | A Sentiment Analysis Method of Capsule Network Based on BiLSTM |
| title_short | A Sentiment Analysis Method of Capsule Network Based on BiLSTM |
| title_sort | sentiment analysis method of capsule network based on bilstm |
| topic | Sentiment analysis neural network capsule network BiLSTM |
| url | https://ieeexplore.ieee.org/document/9007445/ |
| work_keys_str_mv | AT yongfengdong asentimentanalysismethodofcapsulenetworkbasedonbilstm AT yufu asentimentanalysismethodofcapsulenetworkbasedonbilstm AT liqinwang asentimentanalysismethodofcapsulenetworkbasedonbilstm AT yunliangchen asentimentanalysismethodofcapsulenetworkbasedonbilstm AT yaodong asentimentanalysismethodofcapsulenetworkbasedonbilstm AT jianxinli asentimentanalysismethodofcapsulenetworkbasedonbilstm AT yongfengdong sentimentanalysismethodofcapsulenetworkbasedonbilstm AT yufu sentimentanalysismethodofcapsulenetworkbasedonbilstm AT liqinwang sentimentanalysismethodofcapsulenetworkbasedonbilstm AT yunliangchen sentimentanalysismethodofcapsulenetworkbasedonbilstm AT yaodong sentimentanalysismethodofcapsulenetworkbasedonbilstm AT jianxinli sentimentanalysismethodofcapsulenetworkbasedonbilstm |