Sentiment Analysis during the third wave of COVID-19outbreak: A Twitter case study

The disastrous pandemic Covid19 affected all around the world, for the last two years. It has also brought challenges for survival, health, education, and earning a livelihood, such issues brought new trends in society, it adopted social media for communication, work from the home, and online learning platforms helped people to support and fulfill their necessary activities. This study aims in revealing public sentiment during the third wave of Covid-19 caused due to the Omicron variant of coronavirus around the world. In this research article, tweets on various prevalent hashtags associated with the terms Covid-19, Vaccine, Mask, Online Studies and Work from home are collected for five datasets. Public views and opinions are analyzed on these datasets by applying supervised machine learning algorithms based on the Artificial Immune System (AIS) which is inspired by the remarkable ability of the human immune system for information memorization and self-learning system. A comparative study is done among the AIS algorithms and Non- Artificial Immune system algorithms like Deep learning algorithm, K star, Learning Vector Quantization, Naïve Bayes, One Rule, Random Forest, and Sequential Minimal Optimization algorithms. Apart from accuracy, various factors like sensitivity, time-taken to build the model, F-measure, and precision are also compared among the algorithms. The research concludes that in most cases Artificial Immune system-based algorithms work satisfactorily and present accurate predictions.