As schools transitioned to online learning due to COVID-19, there was a wide range of responses from teachers, parents, and students. Some students found online classes conducive to their learning style while other students struggled to adapt. Through various social media platforms, people have been able to share their varying thoughts and emotions towards virtual learning. As a student who experienced the transition, I found some aspects of online classes beneficial such as the increased scheduling flexability. However, many of my peers and professors have expressed opposing views. Using a dataset of tweets with pre-defined hashtags related to online learning, I was interested to see the distribution of positive and negative feelings across the world. In this analysis, I use text mining techniques to explore sentiment towards online learning by calculating tweet polarities. I also use the topic modeling method Latent Dirichlet Allocation (LDA) to discover common themes across the tweets.
See some of the findings below.
There were more positive tweets about distance learning than negative tweets about distance learning.
Nuetral tweets about distance learning get more retweets than highly negative or positve tweets about distance learning.
On average, positive tweets are more subjective than negative tweets.
Word cloud of the most commom words found in positive tweets
Word cloud of the most common words found in negative tweets
