计算机科学

计算机科学 ›› 2024, Vol. 51 ›› Issue (6): 309-316.doi: 10.11896/jsjkx.230400001

• 人工智能 • 上一篇    下一篇

基于多任务联合训练的长文本多实体情感分析

张昊妍1, 段利国1,2, 王钦晨1, 郜浩1   

  1. 1 太原理工大学计算机科学与技术学院 太原 030024
    2 山西电子科技学院信创产业学院 山西 临汾 041000
  • 收稿日期:2023-04-03 修回日期:2023-07-26 出版日期:2024-06-15 发布日期:2024-06-05
  • 通讯作者: 段利国(m15731271621@163.com)
  • 作者简介:(2294187565@qq.com)
  • 基金资助:
    山西省自然科学基金面上项目(202203021221234)

Long Text Multi-entity Sentiment Analysis Based on Multi-task Joint Training

ZHANG Haoyan1, DUAN Liguo1,2, WANG Qinchen1, GAO Hao1   

  1. 1 College of Computer Science and Technology,Taiyuan University of Technology,Taiyuan 030024,China
    2 School of Information and Innovation Industry,Shanxi Electronic Science and Technology Institute,Linfen,Shanxi 041000,China
  • Received:2023-04-03 Revised:2023-07-26 Online:2024-06-15 Published:2024-06-05
  • About author:ZHANG Haoyan,born in 1999,postgraduate.Her main research interests include text sentiment analysis and so on.
    DUAN Liguo,born in 1970,Ph.D,associate professor,postgraduate supervisor,is a senior member of CCF(No.15823S).His main research interests include natural language processing and so on.
  • Supported by:
    General Project of the Natural Science Foundation of Shanxi Province,China(202203021221234).

PDF (PC)

471

摘要： 多实体情感分析旨在识别文中的核心实体并判断其对应的情感,是目前细粒度情感分析领域的研究热点,对长文本多实体情感分析的研究目前还处于起步阶段。文中提出了一种基于多任务联合训练的长文本多实体情感分析模型(PAM),首先采用TF-IDF算法提取文章中与标题相似的句子,剔除冗余信息以缩短文本长度,通过两个BiLSTM分别进行核心实体识别和情感分析任务的学习,获取各自需要的特征,然后利用融入相对位置信息的多头注意力机制将实体识别任务学习到的知识向情感分析任务传递,实现两个任务的联合学习,最后利用提出的Entity_Extract算法根据实体词在文本中出现的次数和先后位置从模型预测的候选实体中确定核心实体并获取其对应的情感。在搜狐新闻数据集上的实验结果证明了PAM模型的有效性。

关键词: 长文本, 多实体, 细粒度情感分析, 多任务学习

Abstract: Multi-entity sentiment analysis aims to identify core entities in a text and judge their corresponding sentiment,which is a research hotspot in the field of fine-grained sentiment analysis.However,most existing researches of long text multi-entity sentiment analysis is still in its early stages.This paper proposes a long text multi-entity sentiment analysis model(PAM) based on multi-task joint training.To begin with,the utilization of TF-IDF algorithm for extracting sentences similar to the article title can help eliminate redundant information and reduce the length of text.Subsequently,the adoption of two BiLSTM models for core entity recognition and sentiment analysis tasks respectively enables the acquisition of necessary features.Next,multi-head attention mechanism is employed,which is integrated with relative position information,to transfer the knowledge gained from entity recognition task to sentiment analysis task,thus enabling joint learning of the two tasks.Finally,the proposed Entity_Extract algorithm is used to identify core entities from predicted candidate entities according to the number and position of entities in the text and obtain their corresponding emotions.Experimental results on Sohu news datasets demonstrate the effectiveness of PAM model.

Key words: Long text , Multi-entity, Fine-grained sentiment analysis, Multi-task learning

中图分类号: 

  • TP391
[1]WANG H Y,TAO W,YU L Y,et al.Review of fine-grainedsentiment analysis[J].Journal of Henan Institute of Science and Technology(Natural Science Edition),2021,49(4):67-76.
[2]PHAN M H,OGUNBONA P.Modelling Context and Syntactical Features for Aspect-based Sentiment Analysis[C]//Annual Meeting of the Association for Computational Linguistics.2020:3211-3220.
[3]PUNITHA A,PRABU R,DEVANATHAN P,et al.Aspect-Based Sentiment Analysis[J/OL].https://www.ijfmr.com/research-paper.php?id=3237.
[4]WANG F,LAN M,WANG W.Towards a One-stop Solution to Both Aspect Extraction and Sentiment Analysis Tasks with Neural Multi-task Learning[C]//International Joint Conference on Neural Network.2018:1-8.
[5]LECUN Y,BOTTOU L,BENGIO Y,et al.Gradient-BasedLearning Applied to Document Recognition[J].Proceedings of the IEEE,1998,86(11):2278-2324.
[6]HOCHREITER S,SCHMIDHUBER J.Long Short-Term Me-mory[J].Neural Computation,1997,9(8):1735-1780.
[7]GAN C,WANG L,ZHANG Z.Multi-entity sentiment analysis using self-attention based hierarchical dilated convolutional neural network[J].Future Generation Computer Systems,2020,112:116-125.
[8]LI X,BING L,LI P,et al.A Unified Model for Opinion Target Extraction and Target Sentiment Prediction[C]//AAAI Confe-rence on Artificial Intelligence.2019:6714-6721.
[9]CHEN M S,HUA Q R,MAO Y J,et al.An Interactive Lear-ning Network That Maintains Sentiment Consistency in End-to-End Aspect-Based Sentiment Analysis[J].Applied Sciences－Basel,2023,13(16):9327.
[10]LUO H,JI L,LI T,et al.GRACE - Gradient Harmonized and Cascaded Labeling for Aspect-based Sentiment Analysis[C]//Conference on Empirical Methods in Natural Language Proces-sing.2020:54-64.
[11]LI B,LIU Y,WANG X.Gradient Harmonized Single-stage Detector[C]//AAAI Conference on Artificial Intelligence.2019:8577-8584.
[12]HUANG J,CUI Y P,WANG S.Adaptive Local Context andSyntactic Feature Modeling for Aspect-Based Sentiment Analysis[J].Applied Sciences－Basel,2023,13(1):603-603.
[13]HU M,PENG Y,HUANG Z,et al.Open-Domain Targeted Sentiment Analysis via Span-Based Extraction and Classification[C]//Annual Meeting of the Association for Computational Linguistics.2019:537-546.
[14]LIN P,YANG M.A Shared-Private Representation Model with Coarse-to-Fine Extraction for Target Sentiment Analysis[C]//Empirical Methods in Natural Language Processing.2020:4280-4289.
[15]LIANG Y,MENG F,ZHANG J,et al.A dependency syntactic knowledge augmented interactive architecture for end-to-end aspect-based sentiment analysis[J].Neurocomputing,2021,454:291-302.
[16]LUO H,LI T,LIU B,et al.DOER:Dual Cross-Shared RNN For Aspect Term-Polarity Co-Extraction[C]//Annual Meeting of the Association for Computational Linguistics.2019:591-601.
[17]LV Y,WEI F,ZHENG Y,et al.A span-based model for aspect terms extraction and aspect sentiment classification[J].Neural Computing and Applications,2021,33(8):3769-3779.
[18]CHEN W,DU J,ZHANG Z,et al.A Hierarchical Interactive Network for Joint Span-based Aspect-Sentiment Analysis[C]//International Conference on Computational Linguistics.2022:7013-7019.
[19]MISRA I,SHRIVASTAVA A,GUPTA A,et al.Cross-stitch Networks for Multi-task Learning[C]//Computer Vision and Pattern Recognition.2016:3994-400.
[20]CHEN Z,QIAN T.Relation-Aware Collaborative Learning for Unified Aspect-Based Sentiment Analysis[C]//Annual Meeting of the Association for Computational Linguistics.2020:3685-3694.
[21]LIANG Y,MENG F,ZHANG J,et al.An Iterative Multi-Knowledge Transfer Network for Aspect-Based Sentiment Analysis[C]//Empirical Methods in Natural Language Proces-sing.2021:1768-1780.
[22]DEVLIN J,CHANG M,LEE K,et al.BERT:Pre-training ofDeep Bidirectional Transformers for Language Understanding[C]//North American Chapter of the Association for Computational Linguistics.2019.
[23]ZHANG C,REN L,MA F,et al.Structural Bias for Aspect Sentiment Triplet Extraction[C]//International Conference on Computational Linguistics.2022:6736-6745.
[24]ZHANG J,DUAN L G,LI A P,et al.Fine-grained SentimentAnalysis Based on Combination of Attention and Gated Mechanism[J].Computer Science,2021,48(8):226-233.
[25]LI X,BING L,ZHANG W,et al.Exploiting BERT for End-to-End Aspect-based Sentiment Analysis[J].arXiv:1910.00883,2019.
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