A Study of Sentiment Analysis Algorithms for Agricultural Product Reviews Based on Improved BERT Model

被引:21
作者
Cao, Ying [1 ,2 ]
Sun, Zhexing [2 ]
Li, Ling [1 ]
Mo, Weinan [1 ,2 ]
机构
[1] Cent South Univ Forestry & Technol, Coll Mech & Elect Engn, Changsha 410004, Peoples R China
[2] Kunming Univ Sci & Technol, Fac Modern Agr Engn, Kunming 650500, Yunnan, Peoples R China
来源
SYMMETRY-BASEL | 2022年 / 14卷 / 08期
关键词
sentiment analysis; deep neural network; BERT; S-LSTM;
D O I
10.3390/sym14081604
中图分类号
O [数理科学和化学]; P [天文学、地球科学]; Q [生物科学]; N [自然科学总论];
学科分类号
07 ; 0710 ; 09 ;
摘要
With the rise of mobile social networks, an increasing number of consumers are shopping through Internet platforms. The information asymmetry between consumers and producers has caused producers to misjudge the positioning of agricultural products in the market and damaged the interests of consumers. This imbalance between supply and demand is detrimental to the development of the agricultural market. Sentiment tendency analysis of after-sale reviews of agricultural products on the Internet could effectively help consumers evaluate the quality of agricultural products and help enterprises optimize and upgrade their products. Targeting problems such as non-standard expressions and sparse features in agricultural product reviews, this paper proposes a sentiment analysis algorithm based on an improved Bidirectional Encoder Representations from Transformers (BERT) model with symmetrical structure to obtain sentence-level feature vectors of agricultural product evaluations containing complete semantic information. Specifically, we propose a recognition method based on speech rules to identify the emotional tendencies of consumers when evaluating agricultural products and extract consumer demand for agricultural product attributes from online reviews. Our results showed that the F1 value of the trained model reached 89.86% on the test set, which is an increase of 7.05 compared with that of the original BERT model. The agricultural evaluation classification algorithm proposed in this paper could efficiently determine the emotion expressed by the text, which helps to further analyze network evaluation data, extract effective information, and realize the visualization of emotion.
引用
收藏
页数:17
相关论文
共 39 条
[1]   Feature selection and ensemble construction: A two-step method for aspect based sentiment analysis [J].
Akhtar, Md Shad ;
Gupta, Deepak ;
Ekbal, Asif ;
Bhattacharyya, Pushpak .
KNOWLEDGE-BASED SYSTEMS, 2017, 125 :116-135
[2]  
[Anonymous], 2005, P C HUM LANG PROC VA
[3]   CitEnergy : A BERT based model to analyse Citizens' Energy-Tweets [J].
Bedi, Jatin ;
Toshniwal, Durga .
SUSTAINABLE CITIES AND SOCIETY, 2022, 80
[4]   A Payload Based Malicious HTTP Traffic Detection Method Using Transfer Semi-Supervised Learning [J].
Chen, Tieming ;
Chen, Yunpeng ;
Lv, Mingqi ;
He, Gongxun ;
Zhu, Tiantian ;
Wang, Ting ;
Weng, Zhengqiu .
APPLIED SCIENCES-BASEL, 2021, 11 (16)
[5]   Combined kNN Classification and Hierarchical Similarity Hash for Fast Malware Detection [J].
Choi, Sunoh .
APPLIED SCIENCES-BASEL, 2020, 10 (15)
[6]   Sentiment detection in social networks and in collaborative learning environments [J].
Colace, Francesco ;
Casaburi, Luca ;
De Santo, Massimo ;
Greco, Luca .
COMPUTERS IN HUMAN BEHAVIOR, 2015, 51 :1061-1067
[7]   Company event popularity for financial markets using Twitter and sentiment analysis [J].
Daniel, Mariana ;
Neves, Rui Ferreira ;
Horta, Nuno .
EXPERT SYSTEMS WITH APPLICATIONS, 2017, 71 :111-124
[8]   Deep Learning: Methods and Applications [J].
Deng, Li ;
Yu, Dong .
FOUNDATIONS AND TRENDS IN SIGNAL PROCESSING, 2013, 7 (3-4) :I-387
[9]  
Devlin Jacob, 2018, ANN C N AM CHAPTER A
[10]  
Feng J Y, 2021, Trans Chin Soc Agricult Mach, V52, P504