On March 2, from 1-4pm, David Mimno from Princeton University will be offering a hands-on workshop on the topic modeling software MALLET. MALLET is a toolkit for analyzing collections of text documents. In this workshop we will focus on one popular method, statistical topic modeling. We will begin with a description of how to represent text documents as data, but the bulk of the workshop will be a hands-on demonstration of how to use topic models to find patterns in a text corpus. We will cover four phases of modeling: preparing data and selecting a vocabulary, running models, analyzing results in the context of additional variables like time and document tags, and diagnosing problems in model fit. Participants wishing to follow along should bring a laptop with the java runtime environment installed.
