Position-Wise Gated Res2Net-Based Convolutional Network with Selective Fusing for Sentiment Analysis
Sentiment analysis (SA) is an important task in natural language processing in which convolutional neural networks (CNNs) have been successfully applied. However, most existing CNNs can only extract predefined, fixed-scale sentiment features and cannot synthesize flexible, multi-scale sentiment feat...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-04-01
|
| Series: | Entropy |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1099-4300/25/5/740 |
| _version_ | 1797600225199652864 |
|---|---|
| author | Jinfeng Zhou Xiaoqin Zeng Yang Zou Haoran Zhu |
| author_facet | Jinfeng Zhou Xiaoqin Zeng Yang Zou Haoran Zhu |
| author_sort | Jinfeng Zhou |
| collection | DOAJ |
| description | Sentiment analysis (SA) is an important task in natural language processing in which convolutional neural networks (CNNs) have been successfully applied. However, most existing CNNs can only extract predefined, fixed-scale sentiment features and cannot synthesize flexible, multi-scale sentiment features. Moreover, these models’ convolutional and pooling layers gradually lose local detailed information. In this study, a new CNN model based on residual network technology and attention mechanisms is proposed. This model exploits more abundant multi-scale sentiment features and addresses the loss of locally detailed information to enhance the accuracy of sentiment classification. It is primarily composed of a position-wise gated Res2Net (PG-Res2Net) module and a selective fusing module. The PG-Res2Net module can adaptively learn multi-scale sentiment features over a large range using multi-way convolution, residual-like connections, and position-wise gates. The selective fusing module is developed to fully reuse and selectively fuse these features for prediction. The proposed model was evaluated using five baseline datasets. The experimental results demonstrate that the proposed model surpassed the other models in performance. In the best case, the model outperforms the other models by up to 1.2%. Ablation studies and visualizations further revealed the model’s ability to extract and fuse multi-scale sentiment features. |
| first_indexed | 2024-03-11T03:45:27Z |
| format | Article |
| id | doaj.art-21a716e0161d4e29a382376ec028595f |
| institution | Directory Open Access Journal |
| issn | 1099-4300 |
| language | English |
| last_indexed | 2024-03-11T03:45:27Z |
| publishDate | 2023-04-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Entropy |
| spelling | doaj.art-21a716e0161d4e29a382376ec028595f2023-11-18T01:15:43ZengMDPI AGEntropy1099-43002023-04-0125574010.3390/e25050740Position-Wise Gated Res2Net-Based Convolutional Network with Selective Fusing for Sentiment AnalysisJinfeng Zhou0Xiaoqin Zeng1Yang Zou2Haoran Zhu3College of Computer and Information, Hohai University, Nanjing 210098, ChinaCollege of Computer and Information, Hohai University, Nanjing 210098, ChinaCollege of Computer and Information, Hohai University, Nanjing 210098, ChinaCollege of Computer and Information, Hohai University, Nanjing 210098, ChinaSentiment analysis (SA) is an important task in natural language processing in which convolutional neural networks (CNNs) have been successfully applied. However, most existing CNNs can only extract predefined, fixed-scale sentiment features and cannot synthesize flexible, multi-scale sentiment features. Moreover, these models’ convolutional and pooling layers gradually lose local detailed information. In this study, a new CNN model based on residual network technology and attention mechanisms is proposed. This model exploits more abundant multi-scale sentiment features and addresses the loss of locally detailed information to enhance the accuracy of sentiment classification. It is primarily composed of a position-wise gated Res2Net (PG-Res2Net) module and a selective fusing module. The PG-Res2Net module can adaptively learn multi-scale sentiment features over a large range using multi-way convolution, residual-like connections, and position-wise gates. The selective fusing module is developed to fully reuse and selectively fuse these features for prediction. The proposed model was evaluated using five baseline datasets. The experimental results demonstrate that the proposed model surpassed the other models in performance. In the best case, the model outperforms the other models by up to 1.2%. Ablation studies and visualizations further revealed the model’s ability to extract and fuse multi-scale sentiment features.https://www.mdpi.com/1099-4300/25/5/740sentiment analysisdeep neural networksconvolutional neural networkResNetRes2Net |
| spellingShingle | Jinfeng Zhou Xiaoqin Zeng Yang Zou Haoran Zhu Position-Wise Gated Res2Net-Based Convolutional Network with Selective Fusing for Sentiment Analysis Entropy sentiment analysis deep neural networks convolutional neural network ResNet Res2Net |
| title | Position-Wise Gated Res2Net-Based Convolutional Network with Selective Fusing for Sentiment Analysis |
| title_full | Position-Wise Gated Res2Net-Based Convolutional Network with Selective Fusing for Sentiment Analysis |
| title_fullStr | Position-Wise Gated Res2Net-Based Convolutional Network with Selective Fusing for Sentiment Analysis |
| title_full_unstemmed | Position-Wise Gated Res2Net-Based Convolutional Network with Selective Fusing for Sentiment Analysis |
| title_short | Position-Wise Gated Res2Net-Based Convolutional Network with Selective Fusing for Sentiment Analysis |
| title_sort | position wise gated res2net based convolutional network with selective fusing for sentiment analysis |
| topic | sentiment analysis deep neural networks convolutional neural network ResNet Res2Net |
| url | https://www.mdpi.com/1099-4300/25/5/740 |
| work_keys_str_mv | AT jinfengzhou positionwisegatedres2netbasedconvolutionalnetworkwithselectivefusingforsentimentanalysis AT xiaoqinzeng positionwisegatedres2netbasedconvolutionalnetworkwithselectivefusingforsentimentanalysis AT yangzou positionwisegatedres2netbasedconvolutionalnetworkwithselectivefusingforsentimentanalysis AT haoranzhu positionwisegatedres2netbasedconvolutionalnetworkwithselectivefusingforsentimentanalysis |