Combining BERT and CNN for Sentiment Analysis A Case Study on COVID-19

被引:1
作者
Kumar, Gunjan [1 ]
Agrawal, Renuka [1 ]
Sharma, Kanhaiya [1 ]
Gundalwar, Pravin Ramesh [2 ]
Kazi, Aqsa [1 ]
Agrawal, Pratyush [3 ]
Tomar, Manjusha [4 ]
Salagrama, Shailaja [5 ]
机构
[1] Symbiosis Int, Symbiosis Inst Technol, Dept Comp Sci Engn, Pune, Maharashtra, India
[2] Anurag Univ, Dept Informat Technol, Hyderabad, India
[3] Symbiosis Int, Symbiosis Inst Technol, Dept Artificial Intelligence & Machine Learning, Pune, India
[4] Indira Coll Engn & Management, Basic Engn Dept, Pune, India
[5] Univ Cumberlands, Comp Informat Syst, Williamsburg, KY USA
关键词
Sentiment analysis; COVID-19; BERT; CNN; ensemble model; NLP; transfer learning;
D O I
10.14569/IJACSA.2024.0151069
中图分类号
TP301 [理论、方法];
学科分类号
081202 ;
摘要
research focuses on sentiment analysis to understand public opinion on various topics, with an emphasis on COVID-19 discussions on Twitter. By utilizing state-of-the-art Machine Learning (ML) and Natural Language Processing (NLP) techniques, the study analyzes sentiment data to provide valuable insights. The process begins with data preparation, involving text cleaning and length filtering to optimize the dataset for analysis. Two models are employed: a Bidirectional Encoder Representations from Transformers (BERT)-based Deep Learning (DL) model and a Convolutional Neural Network (CNN). The BERT model leverages transfer learning, demonstrating strong performance in sentiment classification, while the CNN model excels at extracting contextual features from the input text. To further enhance accuracy, an ensemble model integrates predictions from both approaches. The study emphasizes the ensemble technique's value for more precise sentiment analysis. Evaluation metrics, including accuracy, classification reports, and confusion matrices, validate the effectiveness of the proposed models and the ensemble approach. This research contributes to the growing field of social media sentiment analysis, particularly during global health crises like COVID-19, and underscores its potential to aid informed decision making based on public sentiment.
引用
收藏
页码:676 / 686
页数:11
相关论文
共 32 条
[11]  
Hussein Doaa Mohey El-Din Mohamed, 2018, Journal of King Saud University - Engineering Sciences, V30, P330, DOI 10.1016/j.jksues.2016.04.002
[12]   Sentiment Analysis of COVID-19 Tweets by Machine Learning and Deep Learning Classifiers [J].
Jain, Ritanshi ;
Bawa, Seema ;
Sharma, Seemu .
ADVANCES IN DATA AND INFORMATION SCIENCES, 2022, 318 :329-339
[13]   BERT-deep CNN: state of the art for sentiment analysis of COVID-19 tweets [J].
Joloudari, Javad Hassannataj ;
Hussain, Sadiq ;
Nematollahi, Mohammad Ali ;
Bagheri, Rouhollah ;
Fazl, Fatemeh ;
Alizadehsani, Roohallah ;
Lashgari, Reza ;
Talukder, Ashis .
SOCIAL NETWORK ANALYSIS AND MINING, 2023, 13 (01)
[14]   A Proposed Sentiment Analysis Deep Learning Algorithm for Analyzing COVID-19 Tweets [J].
Kaur, Harleen ;
Ahsaan, Shafqat Ul ;
Alankar, Bhavya ;
Chang, Victor .
INFORMATION SYSTEMS FRONTIERS, 2021, 23 (06) :1417-1429
[15]   The power of ensemble learning in sentiment analysis [J].
Kazmaier, Jacqueline ;
Vuuren, Jan H. van .
EXPERT SYSTEMS WITH APPLICATIONS, 2022, 187
[16]   CNN for situations understanding based on sentiment analysis of twitter data [J].
Liao, Shiyang ;
Wang, Junbo ;
Yu, Ruiyun ;
Sato, Koichi ;
Cheng, Zixue .
8TH INTERNATIONAL CONFERENCE ON ADVANCES IN INFORMATION TECHNOLOGY, 2017, 111 :376-381
[17]   Sentiment analysis for e-commerce product reviews by deep learning model of Bert-BiGRU-Softmax [J].
Liu, Yi ;
Lu, Jiahuan ;
Yang, Jie ;
Mao, Feng .
MATHEMATICAL BIOSCIENCES AND ENGINEERING, 2020, 17 (06) :7819-7837
[18]  
Mansoor M., 2020, ARXIV, DOI DOI 10.48550/ARXIV.2010.14234
[19]  
Minaee S, 2019, ARXIV
[20]   COVIDSenti: A Large-Scale Benchmark Twitter Data Set for COVID-19 Sentiment Analysis [J].
Naseem, Usman ;
Razzak, Imran ;
Khushi, Matloob ;
Eklund, Peter W. ;
Kim, Jinman .
IEEE TRANSACTIONS ON COMPUTATIONAL SOCIAL SYSTEMS, 2021, 8 (04) :1003-1015