skip to main content
10.1145/3665348.3665358acmotherconferencesArticle/Chapter ViewAbstractPublication PagesgaiisConference Proceedingsconference-collections
research-article

Research on sentiment analysis of Bi-LSTM model combining cosine annealing and multi-head attention

Published: 03 July 2024 Publication History

Abstract

Sentiment analysis technology is widely used in many fields such as network public opinion. It has important research significance in the field of natural language processing. In many network texts, sentiment analysis of each entity has always been an important issue. The traditional Bi-LSTM model has the problem of gradient explosion or excessive correction in the learning rate, which affects the convergence speed and stability of the model. In this paper, a Bi-LSTM optimization model combining progressive preheating and cosine annealing is proposed, which uses word embedding technology, word shape reduction technology and named entity recognition technology. Experiments show that our model can achieve high accuracy on both the training set and the verification set, and the loss is gradually reduced. It shows an accuracy of up to 84.13 % in both positive and negative categories, which is significantly better than the existing baseline model. Through in-depth comparative analysis, the model shows obvious advantages in key performance indicators such as accuracy, recall rate and F1 value.

References

[1]
Tian Shihai, Wang Chunmeng, Yang Wenrui. Research on the early warning mechanism of emergency network public opinion crisis based on ANP and stochastic Petri net [J]. China Management Science, 2023, 31 (10): 215-224. / j.cnki.issn1003-207x. 2020.2433.
[2]
Zhou Qiping, Yang Fang. Research on unsupervised early warning mechanism of network public opinion based on SIS model [J].Information Science, 2019,37 (08) : 51-55.DOI : 10.13833 / j.issn.1007-7634.2019.08.008.
[3]
Wang Ying, Gong Huaping. Research on emotional tendency analysis of big data network public opinion based on emotional dimension-Taking 'Nanchang University Independent Cleaning ' Weibo public opinion event as an example [J].Information Science, 2017,35 (04) : 37-42.DOI : 10.13833 / j.cnki.is.2017.04.007.
[4]
Ye Zhoubo, Luo Shun, Yu Juan. Multi-modal false information detection model combined with social network graph [J / OL]. Computer application research, 1-8 2024-03-09.https://doi.org/10.19734/j.issn.1001-3695.2023.11.0565.
[5]
Zhang Jidong, Zhang Huidi. User sentiment analysis of multimodal emergencies with attention mechanism [J]. Intelligence theory and practice, 2022, 45 (11): 170-177.DOI : 10.16353 / j.cnki.1000-7490.2022.11.022.
[6]
Zhi Shiyao, Wu Zhenru, Chen Tao, etc. Research on sentiment analysis of microblog comments based on ALBERT-BiLSTM-Att [J]. Computer Age, 2022, (02): 19-22.DOI : 10.16644 / j.cnki.cn33-1094 / tp.2022.02.006.
[7]
B. Ramya; A. Jothi. SENTIMENT ANALYSIS USING DEEP LEARNING. International Journal of Emerging Technologies and Innovative Research 2023
[8]
Sreevidya P.; O. V. Ramana Murthy. Sentiment analysis by deep learning approaches. S. Veni TELKOMNIKA Telecommunication, Computing, Electronics and Control 2020
[9]
Rajeev Dixit. Deep Learning for Movie Review Sentiment Analysis. International Journal of Engineering Research in Computer Science and Engineering 2023
[10]
Jiang Cuiqing, Guo Yibo, Liu Yao. Research on the construction method of domain sentiment dictionary based on Chinese social media text [J]. Data analysis and knowledge discovery, 2019, 3 (02): 98-107.
[11]
Zhou Zhi, Wang Chunying, Zhu Jiali. Research on the construction of sentiment dictionary in the field of books based on ultra-short comments [J]. Information theory and practice, 2021, 44 (09): 183-189 + 197. / j.cnki.1000-7490.2021.09.026.
[12]
Deng Ciyun, Yu Guoqing. Research on film review sentiment analysis based on Naive Bayes [J]. Intelligent computer and application, 2023, 13(02): 210-212 + 217.
[13]
Liu Gang, Zhang Weishi. Netizens ' evaluation sentiment analysis based on decision tree [J]. Modern Computer (Professional Edition), 2017, (32): 15-19.
[14]
Yan Duanwu, Yang Xiongfei, Li Tiejun. Sentiment analysis of product reviews based on product feature tree and LSTM model [J]. Information theory and practice, 2019, 42 (12): 134-138. / j.cnki.1000-7490.2019.12.02.
[15]
Pei Songwen, Wang Lulu. Research on text sentiment orientation based on attention mechanism [J]. Computer Engineering and Science, 2019, 41 (02): 343-353.

Index Terms

  1. Research on sentiment analysis of Bi-LSTM model combining cosine annealing and multi-head attention

    Recommendations

    Comments

    Information & Contributors

    Information

    Published In

    cover image ACM Other conferences
    GAIIS '24: Proceedings of the 2024 International Conference on Generative Artificial Intelligence and Information Security
    May 2024
    439 pages
    ISBN:9798400709562
    DOI:10.1145/3665348
    Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. Copyrights for components of this work owned by others than the author(s) must be honored. Abstracting with credit is permitted. To copy otherwise, or republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee. Request permissions from [email protected].

    Publisher

    Association for Computing Machinery

    New York, NY, United States

    Publication History

    Published: 03 July 2024

    Permissions

    Request permissions for this article.

    Check for updates

    Qualifiers

    • Research-article
    • Research
    • Refereed limited

    Conference

    GAIIS 2024

    Contributors

    Other Metrics

    Bibliometrics & Citations

    Bibliometrics

    Article Metrics

    • 0
      Total Citations
    • 20
      Total Downloads
    • Downloads (Last 12 months)20
    • Downloads (Last 6 weeks)4
    Reflects downloads up to 17 Feb 2025

    Other Metrics

    Citations

    View Options

    Login options

    View options

    PDF

    View or Download as a PDF file.

    PDF

    eReader

    View online with eReader.

    eReader

    HTML Format

    View this article in HTML Format.

    HTML Format

    Figures

    Tables

    Media

    Share

    Share

    Share this Publication link

    Share on social media