Librarians manually keyword large datasets to enhance information findability in IHME’s catalog. Automating this process could significantly streamline the cataloging workflow. This project documents data characteristics of IHME’s collection to inform an automatic solution. Research revealed that an auto-keywording tool which could process ALL data types would require large investment into a sophisticated computational linguistics solution. This project delivers a practical solution focused on a few specific data types to immediately save Librarians hours of work per data set. The project also provides documentation as lasting evidence for future development and/or funding of a larger Language Learning Model project.