First Archival Vistas Seminar of the Year on March 20 with Amsterdam-based Archivist Arvid de Raaji

Netherlands archivist Arvid de Raaij is presenting for our first Archival Vistas of the year on Friday, March 20, at 1pm EDT. Arvid de Raaji will be discussing his knowledge and experience with Natural Language Processing techniques in the archival setting: "Curatorial Footprints and Data-Driven Methods for the Description of Archives."

Register now!

What happens when we allow archival material to describe itself?

Recent decades have seen a significant increase in digitization within the archival sector, as well as the widespread adoption of Natural Language Processing techniques (such as Automated Text Recognition, Named Entity Recognition, document classification, etc.) to make digital archival materials more accessible. These developments have gone hand in hand with the realization that the archivist is not merely 'a neutral guardian of old paper', but rather a subjective agent that has to interpret materials in order to make them accessible. Using a simple and transparent information retrieval algorithm (TF-IDF) and a substantial collection of egodocuments from the WWII-period in the Netherlands, this session shows how the application of automated keyword extraction helps us think critically about the role of interpretation in archival description and its consequences. And maybe more importantly: how can we incorporate these ideas into source criticism?