How to Machine-Extract Archival Data: Creating Archivally-Minded Computational Models

Computational approaches to archives have gained traction in recent years. Archivists are using computational technologies to appraise, arrange and describe, and create access to digital records. While computational tools and methods have proven useful in specific cases, they have only recently been adopted by archivists, and the complexities of their implementation in an archival context is only beginning to be understood. This paper discusses a pilot project to implement computational approaches to digitized archival records: The Cybernetics Thought Collective: A History of Science and Technology Portal Project, was a collaborative multi-institutional initiative funded by the National Endowment for the Humanities (US). As an example of using computational methods on archives, this paper seeks to contribute to discussions about how computational archival projects can employ models that are informed by archival practices and that generate trustworthy data.