COVID-19 based on NLP for topic discovery and sentiment classification using BiLSTM recurrent neural network approach
M Balasubramaniam, Saravanakumar, Sridhar, Jenifer Grace Giftlin · AIP conference proceedings · 2024
The World Health Organization (WHO) has declared COVID-19 an epidemic, putting human lives at risk worldwide.Researchers are concerned about the recent development of the COVID-19 virus, which first appeared in animals and birds before infecting and causing human illnesses.It research made use of an automated method for extracting COVID-19-related social media postings and a natural language processing (NLP) technique based on topic modelling to unearth a number of issues about COVID-19 from the perspective of the general population.For sentiment classification of COVID-19 comments, it study utilized a BiLSTM with a recurrent neural network (RNN).COVID-19-Sentiment Classification accuracy of 92.15% was also significantly higher than that of other popular machine learning (ML) methods, as demonstrated in experiments.