Building a Digital Archive of Soviet Periodicals: Tools, Workflows, and AI Agents for Vsesvit

Thousands of Soviet-era journals have been scanned, but a scan isn’t data yet. This hands-on workshop follows one digital periodical project from bibliography to full text and analytic exploration, moving from manual methods to experiments with AI agents.

Our case study is the digital archive of Vsesvit (“The Universe”), a Kyiv journal of world literature in Ukrainian translation (vsesvit.vercel.app). Workshop participants will explore this rich collection and its tools for data analysis – including advanced search, charts, network graphs, and chat – as well as the open-source publishing framework that can be used by other scholars wishing to build a similar resource.

This event is co-hosted by the CDH Slavic Digital Humanities Working Group and PIIRS Program in Russian, East European and Eurasian Studies.

Speakers:

Valeriia Korotkova is a Digital Humanities and data specialist at Stanford University Libraries. She works on preserving cultural collections and making them accessible. She develops Vsesvit: Digital Archive and is a data specialist for the Tamizdat Project, a digital archive of banned and unofficial publications from the Cold War to today.

Georgii Korotkov is a PhD candidate in Slavic Languages and Literatures at Stanford, with a minor in Computer Science. He studies Soviet and Eastern European translation journals as sites of literary and cultural exchange, using NLP, network analysis and full-stack development. His latest article, “Reader-scale Text Mining and Visualization in the Secondary ELA Classroom,” appeared in the Journal of Interactive Technology and Pedagogy.