Visualizing online communities on social media

MozFest 2022 Exhibit

In the MozFest Trustworthy AI Working Group’s project on visualizing communities and subcultures on social media, we set out to create a visualization that captured the dynamic time dependence of interactions between communities on the Internet. The visualization above is a proof-of-concept of the kinds of dynamics and interactions that can be captured using a graph visualization. We focus on two COVID-related hashtags that parent groups have been using on Twitter - #teamreality and #urgencyofnormal. Each node represents a hashtag, and each edge represents the two connected hashtags apprearing in the same Tweet, with the frequency of how often the appear signified by the color and width of the edge. You can use the slider to see the time dependence of this co-occurence. If you hover over the node, you can see all the hashtags that appear with the selected one. You can also click on the legend to isolate edges in different frequency ranges.

To see how the data was retrieved and the visualization was created, see this notebook

Visualization created by:

with special thanks to the working group participants: