Title: Public perception analysis on COVID-19 tweets using hybridised method

Authors: Radha Krishna Jana; Dharmpal Singh; Saikat Maity; Hrithik Paul; Ankush Mallick; Sayani Ghatak; Saurav Mallik; Mingqiang Wang

Addresses: Department of Computer Science and Engineering, JIS University, Kolkata, 700109, India ' Department of Computer Science and Engineering, JIS University, Kolkata, 700109, India ' Department of Computer Science and Engineering, Sister Nivedita University, Kolkata, 700156, India ' Department of Computer Science and Engineering, JIS University, Kolkata, 700109, India ' Department of Computer Science and Engineering, JIS University, Kolkata, 700109, India ' Department of Computer Science and Engineering, JIS University, Kolkata, 700109, India ' Department of Environmental Health, Harvard T H Chan School of Public Health, Boston, MA 02115, USA ' Stanford Cardiovascular Institute, Stanford University, CA 94305, USA

Abstract: One of the most widely used social media sites, now a days, is Twitter. During COVID-19 pandemic, people used Twitter to see the daily news and tweet and retweet something on Twitter. This paper mainly focuses on sentiment analysis using tweets that have been performed during the COVID-19 pandemic. Based on Text (Tweets) and Sentiment (i.e., Positive and Negative), various computational models are used for this work, viz., long short-term memory (LSTM) and simple recurrent neural network (SimpleRNN). LSTM yields the best accuracy (92%). Based on the data, both these models provide meaningful forecasts, but in order to maintain confidence, explainable artificial intelligence (XAI) has been combined with both models. It transforms a 'black box' model into a reliable model, increasing people's impact and level of confidence in the suggested models.

Keywords: machine learning; sentiment analysis; SimpleRNN; simple recurrent neural network; LSTM; long short-term memory; COVID-19; twitter; tweets; XAI; explainable artificial intelligence.

DOI: 10.1504/IJCBDD.2024.139485

International Journal of Computational Biology and Drug Design, 2024 Vol.16 No.1, pp.19 - 41

Received: 23 Aug 2023
Accepted: 13 Feb 2024
Published online: 02 Jul 2024 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article