Emotional Insights: Analysing Sentiments in Online Classes Amidst the COVID-19 Pandemic

Abstract

The world was catastrophically unprepared to deal with the set of circumstances that arose in the year 2020 as a result of the corona virus disease (COVID-19). Governments around the world were struggling and required emergency action in all sectors, including education, the economy, healthcare, and trade. Education was one of the hardest hit sectors, forcing educational institutions to relocate their operations to a virtual platform in order to conduct online classes. Public sentiments provide actionable insights into the issues confronting various segments of the education community. This study looks at the reactions of Twitter users to online education. During this difficult time, collecting and studying relevant tweets will aid in understanding the needs of stakeholders such as teachers, students, and parents. We propose an interactive tweet analysis model in this paper to help governments and administration understand the genuine emotion surrounding online classes. The model is divided into four sections: data collection, data analysis, data visualization, and machine learning. This paper also compares five machine learning algorithms for sentiment analysis, including the K-Neighbors classifier and the Linear Support Vector Machine.