中国科学院机构知识库网格
Chinese Academy of Sciences Institutional Repositories Grid
首页 机构 成果 学者
  
  1. CAS IR Grid
  2. 计算技术研究所
  3. 中国科学院计算技术研究所
  4. 中国科学院计算技术研究所期刊论文
  5. 英文
Research on Aspect-Level Sentiment Analysis Based on Adversarial Training and Dependency Parsing

文献类型：期刊论文

作者Xu, Erfeng1,2; Zhu, Junwu1; Zhang, Luchen3; Wang, Yi1,2; Lin, Wei1,2
刊名ELECTRONICS
出版日期2024-05-01
卷号13期号:10页码:16
关键词multi head attention mechanism dependency syntactic relationships adjacency matrix adversarial training
DOI10.3390/electronics13101993
英文摘要Aspect-level sentiment analysis is used to predict the sentiment polarity of a specific aspect in a sentence. However, most current research cannot fully utilize semantic information, and the models lack robustness. Therefore, this article proposes a model for aspect-level sentiment analysis based on a combination of adversarial training and dependency syntax analysis. First, BERT is used to transform word vectors and construct adjacency matrices with dependency syntactic relationships to better extract semantic dependency relationships and features between sentence components. A multi-head attention mechanism is used to fuse the features of the two parts, simultaneously perform adversarial training on the BERT embedding layer to enhance model robustness, and, finally, to predict emotional polarity. The model was tested on the SemEval 2014 Task 4 dataset. The experimental results showed that, compared with the baseline model, the model achieved significant performance improvement after incorporating adversarial training and dependency syntax relationships.
资助项目National Key Research and Development Program of China
WOS研究方向Computer Science ; Engineering ; Physics
语种英语
WOS记录号WOS:001232935000001
出版者MDPI
源URL[http://119.78.100.204/handle/2XEOYT63/40044]  
专题中国科学院计算技术研究所期刊论文_英文
通讯作者Zhu, Junwu; Zhang, Luchen
作者单位1.Yangzhou Univ, Sch Informat Engn, Yangzhou 225127, Peoples R China
2.Chinese Acad Sci, Inst Comp Technol, Key Lab Intelligent Informat Proc, Beijing 100190, Peoples R China
3.Natl Comp Network Emergency Response Tech Team, Coordinat Ctr China, Beijing, Peoples R China
推荐引用方式
GB/T 7714		 Xu, Erfeng,Zhu, Junwu,Zhang, Luchen,et al. Research on Aspect-Level Sentiment Analysis Based on Adversarial Training and Dependency Parsing[J]. ELECTRONICS,2024,13(10):16.
APA Xu, Erfeng,Zhu, Junwu,Zhang, Luchen,Wang, Yi,&Lin, Wei.(2024).Research on Aspect-Level Sentiment Analysis Based on Adversarial Training and Dependency Parsing.ELECTRONICS,13(10),16.
MLA Xu, Erfeng,et al."Research on Aspect-Level Sentiment Analysis Based on Adversarial Training and Dependency Parsing".ELECTRONICS 13.10(2024):16.

入库方式： OAI收割

来源：计算技术研究所

浏览0
下载0
收藏0
其他版本

除非特别说明，本系统中所有内容都受版权保护，并保留所有权利。