Is anyone here able to provide insight about how these visualizations are used today by humanities / social science researchers?
As an undergrad I helped [1] a German Lit professor run topic models and produce visualizations like this for a big dataset of 18th-century magazines. It was a really fun project to be involved with, and the project generated a few conference papers for the professor, but I got the impression that these sorts of tools didn't have a big impact on his usual research workflow. Our models had a ton of noise, and he still had to do the due diligence of reading every single article in painstaking detail.
I'd be curious to hear more about how people actually use these tools!
As an undergrad I helped [1] a German Lit professor run topic models and produce visualizations like this for a big dataset of 18th-century magazines. It was a really fun project to be involved with, and the project generated a few conference papers for the professor, but I got the impression that these sorts of tools didn't have a big impact on his usual research workflow. Our models had a ton of noise, and he still had to do the due diligence of reading every single article in painstaking detail.
I'd be curious to hear more about how people actually use these tools!
[1] https://benrbray.com/posts/2016/digital-humanities-german-pe...