Research on Sentiment Analysis of Government Short Video Comments based on BERT's Multi Strategy Combination Model

Yucong Zhang

doi:10.54097/fcis.v3i3.8570

Authors

Yucong Zhang

DOI:

https://doi.org/10.54097/fcis.v3i3.8570

Keywords:

BERT, Short Video Comment, Sentiment Analysis

Abstract

In the field of NLP, pre trained models greatly shorten development time, reduce usage difficulty, and improve model robustness. In this paper, the BERT pre training model is used to analyze the text emotion of Tiktok short video comments. Through experimental comparison, it is found that the BERT pre training model is better than CNN and BiLSTM in emotion classification. Based on the BERT model, the multi strategy hybrid method is used to further input the BERT output dynamic word vector into the CNN and BiLSTM models to further extract local features and features with long distance dependency, improve the accuracy of the model, and obtain a good classification effect on the government short video comment dataset, which is better than the single model.

Downloads

Download data is not yet available.

References

Liu Luchuan, Sun Yilu The impact of negative online comments with different emotional intensities on consumer attitudes: an experimental study based on regulatory orientation theory [J] Intelligence Theory and Practice, 2020, 43 (5): 163-169.

Wang Chaoyang, Yu Huilin The Emotional Preference Effect in News Short Video Communication: An Empirical Study Based on the Social Segment of Pear Video [J] News and Communication Review, 2019, 72 (3): 42-55.

Qian Chen, Zhang Tao Analysis of transmission characteristics and influencing factors of AIDS related short videos [J] AIDS and STD in China, 2021, 27 (12): 1394-1399.

Du Yongping, Zhao Xiaozheng, and Pei Bingbing Short text sentiment classification based on CNN-LSTM model [J] Journal of Beijing University of Technology, 2019, 45 (7): 662-670.

Wei Lai, Wang Weijie Emotional Analysis of Bullet Screen Information on Video Websites Based on the Theory of Interactive Ritual Chain - Taking Bilibili Health Science Popularization Video as an Example [J] Intelligence Theory and Practice, 2022, 45 (9): 119-126.

Yang Dashen, Li Shixuan, Cong Yingnan Research on the influencing factors of short video transmission effect of Tiktok reading promotion [J] Library science Research, 2021 (23): 34-44.

Qiu Jiangnan, Ge Yidi The Impact of Social Media Emotions on Information Behavior: A Comparative Study Based on Two Types of Disaster Events [J] Management Science, 2020, 33 (1): 3-15.

Guo Xiuyuan, Xue Shipei, Xu Jiayao Research on the audience engagement of regional promotional videos [J] China Television, 2022 (11): 68-74.

Wang Haiyan, Li Jianhui, Yang Fenglei Overview of Support Vector Machine Theory and Algorithm Research [J] Computer Application Research, 2014, 31 (5): 1281-1286.

Zhang Lei, Xiao Siyao, Yang Zelai, etc Research on Public Opinion Processing Based on Improved KNN Algorithm and SIR Model [J] Computer Simulation, 2021, 38 (5): 477-483.

Su Ying, Zhang Yong, Hu Po, etc Emotional analysis based on the combination of naive Bayes and potential Dirichlet distribution [J] Computer Applications, 2016, 36 (6): 1613-1618.

Wan Yanping, Gu Jiazhen, Zhang Fang Integrating Improved Stacking and Rules for Text Emotional Analysis [J] Small Micro Computer Systems, 2021, 42 (7): 1389-1395.

He Xueqin, Yang Wenzhong, Wu Shouer Slam, et al Integrating syntactic rules and CNN for emotional analysis of tourism reviews [J] Computer Engineering and Design, 2019, 40 (11): 3306-3312.

Shang Rongxuan, Zhang Bin, Mi Jianing An end-to-end aspect level sentiment analysis method for government app comments based on BRNN [J] Data Analysis and Knowledge Discovery, 2022, 6 (1): 364-375.

Zhang Yangsen, Zheng Jia, Huang Gaijuan, et al Weibo sentiment analysis method based on dual attention model [J] Journal of Tsinghua University (Natural Science Edition), 2018, 58 (2): 122-130.

Rehman A U, Malik A K, Raza B, et al. A hybrid CNN-LSTM model for improving accuracy of movie reviews sentiment analysis[J]. Multimedia Tools and Applications, 2019, 78(18): 26597-26613.

Wei J, Liao J, Yang Z, et al. BiLSTM with multi-polarity orthogonal attention for implicit sentiment analysis[J]. Neurocomputing, 2020, 383: 165-173.

Duan Dandan, Tang Jiashan, Wen Yong, et al Chinese short text classification algorithm based on BERT model [J] Computer Engineering, 2021, 47 (1): 79-86.

Li Tiefei, Sheng Long, Wu Di Research on Text Classification Method of BERT-TECNN Model [J] Computer Engineering and Applications, 2021, 57 (18): 186-193.

Chen Zhiqun, Ju Ting Research on Weibo Comment Tendency Analysis Based on BERT and Bidirectional LSTM [J] Intelligence Theory and Practice, 2020, 43 (8): 173-177.

Zhang Mingquan, Zhou Hui, Cao Jingang Research on Double BERT Directed Emotional Text Classification Based on Attention Mechanism [J] Journal of Intelligent Systems, 2022, 17 (6): 1220-1227.