ABSTRACT
Aspect-level sentiment analysis is a sub-task of fine-grained sentiment analysis, which aims to predict the sentiment polarity of specific entities or aspects in the identified text to provide richer sentiment information. In recent years, notable advancements have been achieved in the realm of aspect-level text sentiment analysis research. However, the existing methods still have the problems that the traditional pre-trained model cannot solve the problem of multiple meanings of a word, and the existing attention mechanism model cannot recognize context words as sentiment aspect words accurately. Therefore, an aspect-level text sentiment analysis model based on XLNet-GCN is proposed in this paper. Firstly, XLNet is used to generate dynamic word vector representations of text sequences. Then, the text vectors are inputted into BiLSTM to extract text features, and atop the BiLSTM output, multi-layer graph convolution is executed to acquire aspect-specific features. Finally, the sentiment polarity is computed as the model output. Compared with the classical aspect-specific graph convolutional neural network model (ASGCN), our proposed model has improved the accuracy and F1 value by 0.08%∼1.55% and 0.84%∼2.01% in five benchmark datasets.
- Pang, B. , & Lee, L. . (2008). Opinion mining and sentiment analysis. Foundations and Trends® in Information Retrieval, 2(1–2), 1-135.DOI:10.1561/1500000011Google ScholarDigital Library
- Pontiki M , Galanis D , Papageorgiou H ,et al (2018). SemEval-2016 Task 5: Aspect Based Sentiment Analysis. International Workshop on Semantic Evaluation.DOI:10.18653/v1/S16-1002Google ScholarCross Ref
- Xiaowan, S. , Ying, W. , Xin, W. , & Yudong, S. . (2019). Aspect-based sentiment analysis model based on dual-attention networks. Journal of Computer Research and Development(11).DOI:10.7544/ISSN1000-1239.2019.20180823Google ScholarCross Ref
- Xue, W. , & Li, T. . (2018). Aspect based sentiment analysis with gated convolutional networks.DOI:10.18653/v1/P18-1234Google ScholarCross Ref
- Kipf, T. N. , & Welling, M. . (2016). Semi-supervised classification with graph convolutional networks. 2016.DOI:10.48550/arXiv.1609.02907Google ScholarCross Ref
- Mikolov T , Sutskever I , Chen K ,et al..(2013).Distributed Representations of Words and Phrases and their Compositionality.DOI:10.48550/arXiv.1310.4546Google ScholarCross Ref
- Pennington, J. , Socher, R. , & Manning, C. . (2014). Glove: Global Vectors for Word Representation. Conference on Empirical Methods in Natural Language Processing.DOI:10.3115/v1/D14-1162Google ScholarCross Ref
- Peters, M. , Neumann, M. , Iyyer, M. , Gardner, M. , & Zettlemoyer, L. . (2018). Deep contextualized word representations.DOI:10.18653/v1/N18-1202Google ScholarCross Ref
- Radford A, Narasimhan K, Salimans T, . (2018).Improving language understanding by generative pre-training[J]..https://gluebenchmark.com/leaderboardGoogle Scholar
- Devlin J , Chang M W , Lee K ,et al. . (2018). BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding[J]. DOI:10.48550/arXiv.1810.04805Google ScholarCross Ref
- Yang Z , Dai Z , Yang Y ,et al. . (2019). XLNet: Generalized Autoregressive Pretraining for Language Understanding[J]. 2019.DOI:10.48550/arXiv.1906.08237Google ScholarCross Ref
- Tang, D. , Qin, B. , Feng, X. , & Liu, T. . (2015). Effective lstms for target-dependent sentiment classification. Computer Science.DOI:10.48550/arXiv.1512.01100Google ScholarCross Ref
- Huang, B. , Ou, Y. , & Carley, K. M. . (2018). Aspect level sentiment classification with attention-over-attention neural networks. arXiv e-prints.DOI:10.1007/978-3-319-93372-6_22Google ScholarCross Ref
- Ma, D. , Li, S. , Zhang, X. , & Wang, H. . (2017). Interactive attention networks for aspect-level sentiment classification.DOI:10.24963/ijcai.2017/568Google ScholarCross Ref
- Zhang, C. , Li, Q. , & Song, D. . (2019). Aspect-based sentiment classification with aspect-specific graph convolutional networks.DOI:10.18653/v1/D19-1464Google ScholarCross Ref
Index Terms
- Research on Aspect-based Sentiment Analysis Based on XLNet-GCN
Recommendations
Sentence compression for aspect-based sentiment analysis
Sentiment analysis, which addresses the computational treatment of opinion, sentiment, and subjectivity in text, has received considerable attention in recent years. In contrast to the traditional coarse-grained sentiment analysis tasks, such as ...
Aspect-gated graph convolutional networks for aspect-based sentiment analysis
AbstractAspect-based sentiment analysis aims to predict the sentiment polarity of each specific aspect term in a given sentence. However, the previous models ignore syntactical constraints and long-range sentiment dependencies and mistakenly identify ...
Dependency-enhanced graph convolutional networks for aspect-based sentiment analysis
AbstractAspect-based sentiment analysis aims to extract aspect and opinion terms, and identify the sentiment polarities for such terms. The majority of research has proposed effective methods in individual subtasks, and some multi-task learning models ...
Comments