Transformer-Based Graph Convolutional Network for Sentiment Analysis
Sentiment Analysis is an essential research topic in the field of natural language processing (NLP) and has attracted the attention of many researchers in the last few years. Recently, deep neural network (DNN) models have been used for sentiment analysis tasks, achieving promising results. Although...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-01-01
|
| Series: | Applied Sciences |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2076-3417/12/3/1316 |
| _version_ | 1797489220627988480 |
|---|---|
| author | Barakat AlBadani Ronghua Shi Jian Dong Raeed Al-Sabri Oloulade Babatounde Moctard |
| author_facet | Barakat AlBadani Ronghua Shi Jian Dong Raeed Al-Sabri Oloulade Babatounde Moctard |
| author_sort | Barakat AlBadani |
| collection | DOAJ |
| description | Sentiment Analysis is an essential research topic in the field of natural language processing (NLP) and has attracted the attention of many researchers in the last few years. Recently, deep neural network (DNN) models have been used for sentiment analysis tasks, achieving promising results. Although these models can analyze sequences of arbitrary length, utilizing them in the feature extraction layer of a DNN increases the dimensionality of the feature space. More recently, graph neural networks (GNNs) have achieved a promising performance in different NLP tasks. However, previous models cannot be transferred to a large corpus and neglect the heterogeneity of textual graphs. To overcome these difficulties, we propose a new Transformer-based graph convolutional network for heterogeneous graphs called Sentiment Transformer Graph Convolutional Network (ST-GCN). To the best of our knowledge, this is the first study to model the sentiment corpus as a heterogeneous graph and learn document and word embeddings using the proposed sentiment graph transformer neural network. In addition, our model offers an easy mechanism to fuse node positional information for graph datasets using Laplacian eigenvectors. Extensive experiments on four standard datasets show that our model outperforms the existing state-of-the-art models. |
| first_indexed | 2024-03-10T00:13:23Z |
| format | Article |
| id | doaj.art-bfec69af0a644f3894f65193c9f922b5 |
| institution | Directory Open Access Journal |
| issn | 2076-3417 |
| language | English |
| last_indexed | 2024-03-10T00:13:23Z |
| publishDate | 2022-01-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Applied Sciences |
| spelling | doaj.art-bfec69af0a644f3894f65193c9f922b52023-11-23T15:55:33ZengMDPI AGApplied Sciences2076-34172022-01-01123131610.3390/app12031316Transformer-Based Graph Convolutional Network for Sentiment AnalysisBarakat AlBadani0Ronghua Shi1Jian Dong2Raeed Al-Sabri3Oloulade Babatounde Moctard4School of Computer Science and Engineering, Central South University, Changsha 410083, ChinaSchool of Computer Science and Engineering, Central South University, Changsha 410083, ChinaSchool of Computer Science and Engineering, Central South University, Changsha 410083, ChinaSchool of Computer Science and Engineering, Central South University, Changsha 410083, ChinaSchool of Computer Science and Engineering, Central South University, Changsha 410083, ChinaSentiment Analysis is an essential research topic in the field of natural language processing (NLP) and has attracted the attention of many researchers in the last few years. Recently, deep neural network (DNN) models have been used for sentiment analysis tasks, achieving promising results. Although these models can analyze sequences of arbitrary length, utilizing them in the feature extraction layer of a DNN increases the dimensionality of the feature space. More recently, graph neural networks (GNNs) have achieved a promising performance in different NLP tasks. However, previous models cannot be transferred to a large corpus and neglect the heterogeneity of textual graphs. To overcome these difficulties, we propose a new Transformer-based graph convolutional network for heterogeneous graphs called Sentiment Transformer Graph Convolutional Network (ST-GCN). To the best of our knowledge, this is the first study to model the sentiment corpus as a heterogeneous graph and learn document and word embeddings using the proposed sentiment graph transformer neural network. In addition, our model offers an easy mechanism to fuse node positional information for graph datasets using Laplacian eigenvectors. Extensive experiments on four standard datasets show that our model outperforms the existing state-of-the-art models.https://www.mdpi.com/2076-3417/12/3/1316sentiment analysisgraph neural networkdeep learningNLP transformer |
| spellingShingle | Barakat AlBadani Ronghua Shi Jian Dong Raeed Al-Sabri Oloulade Babatounde Moctard Transformer-Based Graph Convolutional Network for Sentiment Analysis Applied Sciences sentiment analysis graph neural network deep learning NLP transformer |
| title | Transformer-Based Graph Convolutional Network for Sentiment Analysis |
| title_full | Transformer-Based Graph Convolutional Network for Sentiment Analysis |
| title_fullStr | Transformer-Based Graph Convolutional Network for Sentiment Analysis |
| title_full_unstemmed | Transformer-Based Graph Convolutional Network for Sentiment Analysis |
| title_short | Transformer-Based Graph Convolutional Network for Sentiment Analysis |
| title_sort | transformer based graph convolutional network for sentiment analysis |
| topic | sentiment analysis graph neural network deep learning NLP transformer |
| url | https://www.mdpi.com/2076-3417/12/3/1316 |
| work_keys_str_mv | AT barakatalbadani transformerbasedgraphconvolutionalnetworkforsentimentanalysis AT ronghuashi transformerbasedgraphconvolutionalnetworkforsentimentanalysis AT jiandong transformerbasedgraphconvolutionalnetworkforsentimentanalysis AT raeedalsabri transformerbasedgraphconvolutionalnetworkforsentimentanalysis AT olouladebabatoundemoctard transformerbasedgraphconvolutionalnetworkforsentimentanalysis |