Investigating the Impact of COVID-19 on Education by Social Network Mining

TL;DR

This study analyzes social media data related to COVID-19 and education to understand public sentiment and trends, revealing correlations between tweet activity and official case numbers across countries.

Contribution

It introduces a method combining geo-tagging and sentiment analysis on Twitter data to study COVID-19's impact on education globally.

Findings

Correlation between tweet frequency and confirmed cases in several countries

Sentiment analysis reveals shifts in public mood over time

Geographical distribution of COVID-19 related tweets

Abstract

The Covid-19 virus has been one of the most discussed topics on social networks in 2020 and 2021 and has affected the classic educational paradigm, worldwide. In this research, many tweets related to the Covid-19 virus and education are considered and geo-tagged with the help of the GeoNames geographic database, which contains a large number of place names. To detect the feeling of users, sentiment analysis is performed using the RoBERTa language-based model. Finally, we obtain the trends of frequency of total, positive, and negative tweets for countries with a high number of Covid-19 confirmed cases. Investigating the results reveals a correlation between the trends of tweet frequency and the official statistic of confirmed cases for several countries.