Online Learning Communities in Covid-19 Days: Mining Twitter Data

Abstract

As a teacher educator who provides online education during the pandemic process, I wanted to examine what is needed and what is shared in online learning communities. The aim of this research is to reveal the most shared concepts and sentiments in online learning networks using the data obtained from Twitter. In this context, the #edtech hashtag, one of the most shared hashtags, was chosen and a total of 134,101 unique tweets were analyzed in the specified time period with some data mining and sentiment analysis techniques. As a result of these analyzes, the most shared websites, words and bigrams were extracted. In addition, sentiment analysis based on NRC and Bing lexicons was also performed. According to the findings about the most shared websites, it has been seen that, in addition to current learning approaches such as digital learning designs, STEAM, coding, robotics, artificial intelligence and augmented reality, various mathematical calculation applications and websites based on fun math activities are shared. The words learning, student, school, education, online, free, classroom, teacher, join and check are the most used words. The most used bigrams are activity pack, remote learning, pupil activity, online learning and google classroom. Considering the sentiment analysis, it was seen that the most prominent sentiment analysis, it was seen that the most prominent sentiment based on the NRC lexicon was positive, followed by the sentiments of trust, anticipation and joy. According to Bing, positive sentiments were dominant.

Keywords

• : Educational technology
• edtech
• text mining
• sentiment analysis
• COVID-19
• : Educational technology, edtech, text mining, sentiment analysis, COVID-19