Twitter sentiment analysis to understand students' perceptions about online learning during the Covid'19

In 2019, the Covid-19 pandemic led to emergency changes in the educational sector as a precautionary measure to limit the spread of the Covid-19 virus and protect the health and safety of students. Educational institutes couldn't escape this havoc; by April 2020, 189 countries had suspended school, affecting 89 percent of the world's students. Since the epidemic began, online learning has completely taken over the educational industry, leaving students with no choice but to adapt to the brand-new virtual learning environment. Consequently, people turned to social media, such as Twitter, to express their feelings, opinions, and concerns about online learning as an alternative to traditional physical classes. The new online learning platforms, associated technologies, and procedures have been widely discussed on Twitter. In the proposed study, we have presented a systematic approach to analyze the public opinions and perceptions about online learning using Twitter sentiment analysis (TSA) through Twitter's API and term frequency-inverse document frequency (TF-IDF) technique. Further, we classified the sentiments into certain clusters, such as positive, negative, and neutral, using a text mining approach (i.e., lexicons). Moreover, we have uncovered these sentiments and visualized the clusters using visualization techniques such as word clouds and bar charts. Additionally, by using TF-IDF, we measured the strength of words that people use to express their opinions about online schooling and explored to what extent it affects the overall results of our analysis.