The Zoom link for today is: https://berkeley.zoom.us/j/265929778
Join us for two online sessions on text data mining with Gale's extensive primary resource collections in a web-based platform called Digital Scholar Lab. The first session will provide an overview of the platform; the second session will provide hands-on practice with two of the Digital Scholar Lab's tools. Participants are invited to join for one or both sessions. There will be a 15 minute break between the sessions.
Session 1: Computational Text Analysis in the Cloud: The Digital Scholar Lab1:00pm-2:00pm
This session introduces you to the Digital Scholar Lab, a cloud-based research platform created by the Gale publishing company, through which you can build custom-curated datasets for computational text analysis. After compiling and cleaning text corpora from primary source documents available through the Library's subscriptions to Gale databases, you can then perform 6 types of computational text analyses using cloud-based tools. Export visualizations, tabular data, and up to 5,000 documents of OCR text.
We’ll walk through how to search, curate, and manage datasets; use analysis tools; and review resulting visualizations. After attending this session you'll feel comfortable building and cleaning datasets, and you'll understand what kinds of analyses are possible in the Digital Scholar Lab.
Session 2: Ngrams and Topic Modeling in The Digital Scholar Lab2:15pm-3:30pm
You can analyze content in the Digital Scholar Lab using 6 quantitative and qualitative analysis methods. In this session, we’ll explore one quantitative tool (Ngrams) and one qualitative tool (Topic Modeling). We’ll cover what the tools do, how they work, how setting configurations change your outputs, and how they can be used to address research and text cleaning questions. This will be a hands-on workshop, so bring your laptop if possible. No prior experience with text analysis necessary. Consider attending Session 1 before this workshop.
