Sentiment Analysis of COVID-19 Tweets by Machine Learning and Deep Learning Classifiers

Sentiment analysis helps in deciding the emotions of a person that they like to express as neutral, negative, or positive. Many algorithms of machine learning and deep learning exist for analyzing sentiments of people posted on various social media platforms. Twitter, one of the social media platforms, is becoming very helpful to understand the behavior of society at large as people conveyed their views during the COVID-19 pandemic by offering details that have been organized as tweets. This paper aims to study the performance of various classification algorithms that demands an input value and recognizes to which output category it belongs. Six machine learning algorithms, two ensemble algorithms, and four deep learning algorithms have been considered for this work. COVID-19 twitter dataset comprising of two lakh records is reviewed. Accuracy, confusion matrix, recall matrix, precision matrix, recall, precision, and F1 -score parameters have been taken as performance measures.