"What is on this disk?" An Exploration of Natural Language Processing in Archival Appraisal Public Deposited

Downloadable Content

Download PDF
Last Modified
  • May 8, 2019
  • Goodman, Morgan
    • Affiliation: School of Information and Library Science
  • This paper explores current processes in archival appraisal and selection and investigates the potential uses of automation in the processes. Through an exploration of the BitCurator NLP topic modeling tool, bitcurator-nlp-gentm, I evaluated reactions by participants who agreed to an interview and exploration of the tool. I conclude that topic modeling can assist archivists through identification of like-collections and possible duplication within hybrid collections. Outside of appraisal, topic modeling tools may have uses for archival description and arrangement. Researchers and those with subject matter expertise may also benefit from these tools. This paper points to areas where topic modeling is effective and offers suggestions for making NLP and topic modeling more universally practical in archival workflows.
Date of publication
Resource type
  • Lee, Cal
  • Master of Science in Information Science
Academic concentration
  • Library and Information Science
Degree granting institution
  • University of North Carolina at Chapel Hill
Graduation year
  • 2019
Deposit record
  • 07b9ef20-a3fe-477b-b4fb-c0a986496c8b

This work has no parents.