Online learning sentiment analysis during the covid-19 Indonesia pandemic using twitter data

Abstract

In the last week of September 2020, Covid-19 in Indonesia has infected more than 252,000 people. The spread of the virus through physical contact forces all countries to use social distance and physical distance to reduce interactions. On the educational side, the Covid-19 pandemic has had an impact on the prevention of face-to-face teaching and learning activities. All educational institutions must stop the learning process and replace it with online learning. Nadiem Makarim, Minister of Education and Culture of the Republic of Indonesia, then affirmed this online learning by issuing a policy on online learning in case of an emergency spread of the Covid-19 virus. However, this online learning has provoked debate in the community over the preparation of the technology. The goal of this study was to analyze public opinion for online learning during the Covid-19 pandemic in October 2020. The research carried out document-based text mining and feelings on Twitter data analyzed using the Naïve Bayes method. Analysis in that period found 25 percent positive sentiment, 74 percent negative sentiment, and 1 percent neutral sentiment. Several tweets showed that the words ‘stress’ and ‘covid’ were the most commonly spoken in the month.