Understanding topic modeling is, to borrow a phrase from George W. Bush, “the war that never ends”. And as I was re-reading the excellent posts by Matt Jockers, Scott Weingart and Ted Underwood, I realized that I may never have made clear that Comprehending the Digital Humanities was much more about understanding how topic modeling worked than it was about understanding how the Digital Humanities defined itself. The reason I chose a corpus of digital humanities articles was because I felt like it was comprehensible to myself and any potential audience.
While I’m sure the name doesn’t help–perhaps, Comprehending Topic Modeling would have been better–I choose to blame the confusion on my lack of legends, so I dug up my favorite data visualization of topic model output, above, and added one in.