Analyze the evolution of COVID-19 topics

For a practical university project I would like to analyze the evolution of COVID-19 topics (such as Corona-Crisis) based on social media data.


Very interesting and up to date topic ! I was bot thinking about Twitter as a relevant medium towards this Problem
Moritz Ruoff Holzer - Wed, 12/22/2021 - 06:25 :::
2 answers

I did this kind of investigation in my bachelor thesis last year, to investigate the spread of misinformation regarding the pandemic. We also used the Twitter API which is really easy to work with. You can use the endpoints to search for specific hashtags, filter the text in tweets, and so on. They also provide endpoints to see specific trends for topics, and you can see how these trends differ around the world. Good luck with your project!

First I will use the Twitter API and retrieve tweets available for the Covid-19 topic. To identify changes in topics I would write an application in python and use natural language processing to obtain the semantics and evaluate the changes of the words based on a similarity technique (such as cosine similarity.) At the end, I would try to display some target topics in graphics and their contextual changes.