Constellate for Text Analysis

UVA Library currently has a trial of Constellate, a text analysis platform that incorporates scholarly content from JSTOR and others and a cloud-based platform to teach text analysis and data literacy skills. The platform includes no-code options, as well as Constellate Lab, based on Jupyter Lab, that includes Python and RStudio.

What is in Constellate?

Constellate contains text data and a dataset builder with over 38 million documents from the following providers:

• JSTOR
• Portico
• Chronicling America
• Doc South
• South Asia Open Archives
• Reveal Digital

How do I learn with the Skill-Build program?

Part of our trial is the opportunity to enroll in Constellate’s first Skill-Build program. This includes a series of scaffolded live and asynchronous classes over three months, covering topics from Python basics to automated text classification using ChatGPT's API to teaching data literacy. I think one of the strengths of Constellate is the education resources they offer, so I’d love to see if people can take advantage of them during the trial (the Skill-Build program is just for individuals at institutions with subscriptions or in trials). The program started in September. All recordings are available in the Slack channel for each class, so interested learned could catch up there. Reach out to the librarian on this page to ask for the Slack link.

You can learn more about Constellate at their user guide.

Who is this for?

The trial is open to UVA students, faculty, and staff.

When does the trial end?

The trial ends December, 31, 2024.

Feedback requested!

If you test Constellate, please email the librarian on this page with your feedback (positive, negative, or neutral).