计算机科学

计算机科学 ›› 2024, Vol. 51 ›› Issue (11A): 240400018-7.doi: 10.11896/jsjkx.240400018

• 智能计算 • 上一篇    下一篇

基于BERT模型和图注意力网络的方面级情感分析

林煌, 李弼程   

  1. 华侨大学计算机科学与技术学院 福建 厦门 361021
  • 出版日期:2024-11-16 发布日期:2024-11-13
  • 通讯作者: 李弼程(lbclm@163.com)
  • 作者简介:(1209285067@qq.com)
  • 基金资助:
    装备预研教育部联合基金(8091B022150)

Aspect-based Sentiment Analysis Based on BERT Model and Graph Attention Network

LIN Huang, LI Bicheng   

  1. College of Computer Science and Technology,Huaqiao University,Xiamen,Fujian 361021,China
  • Online:2024-11-16 Published:2024-11-13
  • About author:LIN Huang,born in 2000,postgraduate.His main research interests include sentiment analysis in natural language processing,intelligent data management and analysis,and deep learning.
    LI Bicheng,born in 1970,professor,Ph.D supervisor.His main research interests include artificial intelligence,natural language processing,public opinion monitoring and guidance.
  • Supported by:
    Joint Fund of Equipment Pre-research and Ministry of Education(8091B022150).

PDF (PC)

201

摘要： 方面级情感分析是一项细粒度的情感分析任务,旨在对给定文本中的特定方面进行情感极性分析。当前基于语法分析的方法严重依赖于依存树的单一解析结果,并且大部分研究对于语义和语法特征的融合并不充分。因此,提出了一种基于BERT模型和图注意力网络的方面级情感分析方法。该方法能够充分挖掘句子结构中的语义和语法信息,并通过交互注意力机制融合这些信息,从而获得更精确的情感特征。首先,利用BERT预训练模型得到文本的初始化向量,并使用注意力机制对方面词进行全局的语义信息关联,得到文本的语义特征。其次,利用语法解析器构建短语结构图和依存图,并利用图注意力网络对节点信息进行编码,得到文本的语法特征。最后,通过交互注意力机制结合学习到的语义和语法特征,实现了多个视角的融合,从而全面理解方面-观点关系。实验结果表明,所提方法在多个数据集上的ACC值和F1值均优于现有的多个先进方法。

关键词: 方面级情感分析, 语法分析, 图注意力网络, 注意力机制

Abstract: Aspect-based sentiment analysis is a fine-grained sentiment analysis task that aims to analyze the sentiment polarity of specific aspects in a given text.Current syntax-based methods heavily rely on a single parsing result from the dependency tree,and most of the existing research lacks sufficient integration of semantic and syntactic features.Therefore,this paper proposes an aspect-based sentiment analysis approach based on the BERT model and graph attention network.This method can effectively explo the semantic and syntactic information in sentence structures and fuse these information through an interactive attention mechanism to obtain more accurate sentiment features.Firstly,the BERT pre-trained model is utilized to obtain the initial vectors of the text,and an attention mechanism is employed to associate the globalsemantic information of aspect words,resulting in the semantic features of the text.Secondly,a syntax parser is used to construct a phrase structure graph and a dependency graph,and a graph attention network is applied to encode the node information,leading to the syntactic features of the text.Finally,through an interactive attention mechanism,the learned semantic and syntactic features are combined to achieve a comprehensive understanding of aspect-opinion relationships from multiple perspectives.Experimental results show that the proposed method outperforms the existing state-of-the-art methods with ACC and F1 values on multiple datasets.

Key words: Aspect-based sentiment analysis, Syntax analysis, Graph attention network, Attention mechanism

中图分类号: 

  • TP391.1
[1]WANG Y,HUANG M,ZHU X,et al.Attentionbased LSTM for aspect-level sentiment classification[C]//Conference on Empirical Methods in Natural Language Processing.2016:606-615.
[2]TANG D,QIN B,FENG X,et al.Effective lstms for target-dependent sentiment classification[C]//International Conference on Computational Linguistics.2016:3298-3307.
[3]ZENG X Q,HUA X,LIU P S,et al.Emotion Wheel and Lexicon based emotion distribution Label Enhancement Method[J].Chinese Journal of Computers,2021,44:1080-1094.
[4]WU H,ZHANG Z,SHI S,et al.Phrase dependency relational graph attention network for Aspect-based Sentiment Analysis[J].Knowledge-Based Systems,2022,236:247.
[5]WU H,ZHANG Z,WU Q.Exploring syntactic and semanticfeatures for authorship attribution[J].Applied Soft Computing,2021,111:127.
[6]ZHANG M,QIAN T.Convolution over hierarchical syntacticand lexical graphs for aspect level sentiment analysis[C]//Conference on Empirical Methods in Natural Language Processing.2020:3540-3549.
[7]WANG K,SHEN W,YANG Y,et al.Relational graph attention network for aspect-based sentiment analysis[C]//Association for Computational Linguistics.2020:3229-3238.
[8]ZHAO G Y,LV C G,FU G H,et al.Domain Specific Sentiment Words Based Attention Model for Cross-Domain Attribute-Oriented Sentiment Analysis[J].Journal of Chinese Information Processing,2021,35(6):93-102.
[9]WANG Y,HUANG M,ZHU X,et al.Attention-based LSTM for aspect-level sentiment classification[C]//Conference on Empirical Methods in Natural Language Processing.2016:606-615.
[10]CHEN P,SUN Z,BING L,et al.Recurrent attention network on memory for aspect sentiment analysis[C]//Conference on Empirical Methods in Natural Language Processing.2017:452-461.
[11]MA D,LI S,ZHANG X,et al.Interactive attention networks foraspect-level sentiment classification[C]//Conference on International Joint Conference on Artificial Intelligence.2017:4068-4074.
[12]LU Q,ZHU Z,ZHANG D,et al.Interactive rule attention network for aspect-level sentiment analysis[J].IEEE Access,2020,8:52505-52516.
[13]XU H,LIU B,SHU L,et al.BERT post-training for review reading comprehension and aspect-based sentiment analysis[C]//Annual Conference of the North American Chapter of the Association for Computational Linguistics.2019:2324-2335.
[14]SONG Y,WANG J,JIANG T,et al.Attentional encoder network for targeted sentiment classification[C]//International Conference on Artificial Neural Networks.2019:93-103.
[15]ZHANG K,ZHANG K,ZHANG M,et al.Incorporating dynamicsemantics into pre-trained language model for aspect-based sentiment analysis[C]//Annual Meeting of the Association for Computational Linguistics.2022:3599-3610.
[16]YAN H,DAI J,QIU X,et al.A unified generative framework for aspect-based sentiment analysis[C]//International Joint Conference on Natural Language Processing.2021:2416-2429.
[17]ZHANG C,LI Q,SONG D.Aspect-based sentiment classifica-tion with aspect-specific graph convolutional networks[C]//Conference on Empirical Methods in Natural Language Proces-sing.2019:4568-4578.
[18]LI R,CHEN H,FENG F,et al.Dual graph convolutional networks for aspect-based sentiment analysis[C]//International Joint Conference on Natural Language Processing.2021:6319-6329.
[19]LIANG B,SU H,GUI L,et al.Aspect-based sentiment analysis via affective knowledge enhanced graph convolutional networks[J].Knowledge-Based Systems,2022,235:1-11.
[20]CHENC,TENG Z,WANG Z,et al.Discrete opinion tree induction for aspect-based sentiment analysis[C]//Annual Meeting of the Association for Computational Linguistics.2022:2051-2064.
[21]ZHAO X,PENG H,DAI Q,et al.Rdgcn:Reinforced dependency graph convolutional network for aspect-based sentiment analysis[C]//International Conference on Web Search and Data Mi-ning.2024:976-984.
[22]ZHAO X,ZHOU Y,XU X,et al.Extensible Multi-GranularityFusion Network for Aspect-based Sentiment Analysis[J].ar-Xiv:2402.07787,2024.
[23]TU K W,LI J.A Survey of Recent Developments in Syntactic Parsing[J].Journal of Chinese Information Processing,2020,34(7):30-41.
[24]CHAI W.A Review of the Constituency Parsing[J].Computer Knowledge and Technology,2020,16(16):3-10.
[25]MRINI K,DERNONCOURT F,BUI T,et al.Rethinking self-attention:An interpretable self-attentive encoder-decoder parser[C]//Conference on Empirical Methods in Natural Language Processing.2020:731-742.
[26]HERCIG T,BRYCHCÍN T,SVOBODA L,et al.Uwb at semeval-2016 task 5:Aspect based sentiment analysis[C]//International Workshop on Semantic Evaluation.2016:342-349.
[27]DONG L,WEI F,TAN C,et al.Adaptive recursive neural network for target-dependent twitter sentiment classification[C]//Annual Meeting of the Association for Computational Linguistics(volume 2:Short papers).2014:49-54.
[28]JIANG Q,CHEN L,XU R,et al.A challenge dataset and effective models for aspect-based sentiment analysis[C]//Conference on Empirical Methods in Natural Language Processing.2019:6280-6285.
[29]FAN F,FENG Y,ZHAO D.Multi-grained attention network for aspect-level sentiment classification[C]//Conference on Empirical Methods in Natural Language Processing.2018:3433-3442.
[30]CHEN P,SUN Z,BING L,et al.Recurrent attention network on memory for aspect sentiment analysis[C]//Conference on Empirical Methods in Natural Language Processing.2017:452-461.
[31]SUN K,ZHANG R,MENSAH S,et al.Aspect-level sentiment analysis via convolution over dependency tree[C]//Conference on Empirical Methods in Natural Language Processing.2019:5679-5688.
[32]MA M,NA S,WANG H.AEGCN:An autoencoder-constrained graph convolutional network[J].Neurocomputing,2021,432:21-31.
[33]ZENG J,LIU T,JIA W,et al.Relation construction for aspect-level sentiment classification[J].Information Sciences,2022,586:209-223.
Viewed
Full text


Abstract

Cited
  Shared   
  Discussed   
No Suggested Reading articles found!
渝ICP备16013121号-4
地址：重庆市渝北区洪湖西路18号    邮编：401121  
电话：023-63500828（编辑部） 023-67039612（会议/专题） 023-67039623（财务/发票）
E-mail：jsjkx12@163.com
本系统由北京玛格泰克科技发展有限公司设计开发