This workshop explores two experimental approaches to using AI in non-Latin script cataloging, with a focus on Middle Eastern materials. In the first presentation, “Resolving Non-Latin Script Cataloging Issues with GenAI and Python: AI-built Unicode Decomposer,” Yasha Razizadeh introduces a Python-based tool developed with generative AI to assist in the decomposition and normalization of non-Latin script metadata.
The second presentation, “AI in Cataloging: Experimenting with Middle Eastern Materials,” is presented by Amal Morsy and Ayman Osman. Aimed at libraries working with non-Latin script collections, it focuses on helping librarians streamline cataloging workflows and providing effective tools for institutions without specialized Arabic catalogers. The session will include a live demonstration of how AI tools like ChatGPT can support cataloging Arabic materials, benefiting both experienced and early-career catalogers by improving efficiency, reducing workload, and enhancing consistency. Practical AI-driven cataloging approaches will be showcased to make the session valuable and accessible to all catalogers, regardless of language expertise.
Presenters:
- Yasha Razizadeh, Metadata Librarian for Non-Latin Script Resources, New York University
- Amal Morsy, Arabic Librarian, Middle East Section, Library of Congress
- Ayman Osman, Rare Materials Cataloger and AI Specialist, University of Florida Libraries