Special CIRSS Seminar: David Mimno

iSchool Research Fellow David Mimno will give the talk, "Computational ethnography: on the convergence between machine learning and grounded theory."

Abstract: Work on machine learning algorithms has become incredibly popular in the past few years, but there has been much less work studying the context of machine learning. Who is using these algorithms, and what do those people want from them? Using the example of survey responses from people voluntarily leaving Facebook, I will show that a machine learning tool (statistical topic modeling) turns out to have remarkable similarities to an ethnographic method (grounded theory). I will then describe the properties of algorithms that appear to explain these connections. By connecting existing research practices to new technical approaches we not only provide useful tools and methodologies, but we also gain more insight into why algorithms are more or less useful in specific contexts.

Mimno is an assistant professor in the Department of Information Science at Cornell University. He holds a PhD from UMass Amherst and was previously the head programmer at the Perseus Project at Tufts and a researcher at Princeton University. His work is supported by the Sloan foundation and the NSF.