In the era of the COVID-19 pandemic, the surge in information sharing on social media, particularly Twitter, necessitates a nuanced understanding of sentiments. Conventional sentiment analysis methods face challenges in capturing the evolving discourse's contextual nuances. This study introduces a novel approach, employing BERT-RCNN for sentiment classification of COVID-19-related tweets. BERT's bidirectional contextual insights combined with RCNN's feature extraction enhance our model's accuracy. The labels 'Neutral,' 'Positive,' and 'Negative' provide a nuanced emotional analysis. Our methodology overcomes traditional limitations, offering a context-aware sentiment analysis. By leveraging BERT-RCNN, this research contributes to a deeper understanding of public sentiments during the pandemic, addressing evolving challenges in sentiment classification.
Xiong, J., Feng, M., Wang, X., Jiang, C., Zhang, N., & Zhao, Z. (2024). Decoding sentiments: Enhancing covid-19 tweet analysis through bert-rcnn fusion. Journal of Theory and Practice of Engineering Science, 4(01), 86-93.
@article{xiong2024decoding,
title={Decoding sentiments: Enhancing covid-19 tweet analysis through bert-rcnn fusion},
author={Xiong, Jize and Feng, Mingyang and Wang, Xiaosong and Jiang, Chufeng and Zhang, Ning and Zhao, Zhiming},
journal={Journal of Theory and Practice of Engineering Science},
volume={4},
number={01},
pages={86--93},
year={2024}
}