Tutorial: Generating and Visualizing Topic Models with Tethne and MALLET
By Erick Peirson.
Tethne provides a variety of methods for working with text corpora and the output of modeling tools like MALLET. This tutorial focuses on parsing, modeling, and visualizing a Latent Dirichlet Allocation topic model, using data from the JSTOR Data-for-Research portal.
In this tutorial, we will use Tethne to prepare a JSTOR DfR corpus for topic modeling in MALLET, and then use the results to generate a semantic network like the one shown below.
In this visualization, words are connected if they are associated with the same topic; the heavier the edge, the more strongly those words are associated with that topic. Each topic is represented by a different color. The size of each word indicates the structural importance (betweenness centrality) of that word in the semantic network.
This tutorial was developed for the course Introduction to Digital & Computational Methods in the Humanities (HPS), created and taught by Julia Damerow and Erick Peirson.