Register for this workshop at the link on the Constellate website: https://constellate.org/events/learn-about-text-analysis-may-8

Spend three hours with Amy and other interested researchers learning about different text analysis techniques at a measured pace.  This is a stress free, exploratory, fun session – we will keep it light and microphones may stay on.  If you do not have time for the whole session, you are welcome to attend the first hour only or join us for one of the other times at which this session will be offered in April & May 2024.

Our agenda:

Learn about text analysis methods (45 minutes). Using the new Constellate key concept tutorials, we will consider all the things you can do by counting words.

Building datasets (15 minutes). Next, we are going to build our own datasets of either journal articles, books and book chapters, historic American newspapers issues, or primary sources in the Constellate dataset builder.

Break

Analyze our datasets (1.5 hours). We will end by doing some work with the datasets we just built using Constellate's pre-built dataset files and notebooks, including:

• what can you do by downloading dataset files and working with them in software you already have?
• pre-processing (do you really want to include all that front matter?)
• word counts (the CSV file Constellate provides of word counts is really large, let's use code to narrow it down a bit)
• topic modeling (can we find latent topics in each of our datasets?)

This class is open to users at participating and trialing institutions (non-participants may attend the session, but you will not be able to access the tutorials or lab.)  No materials or software are required, however we will talk about opening dataset files in OpenRefine and Tableau (OpenRefine is freely available and you may have access to Tableau, a data visualization tool, through your institution.)  Register to attend in person or access the recording asynchronously.