Abstract: Researchers across the disciplines had several years to consider what it means to have computational access to one million books. But what about maps? Though there is no global count, we’re likely nearing the million mark for cartographic materials in library and archival digital collections. With so many images being scanned, researchers can now test and improve automatic methods for generating data from large collections of maps. The visual turn in digital humanities emphasizes the opportunities of machine learning with the lens of source criticism. So, what kinds of questions might we ask of thousands of maps? How can we connect maps with other big data? In this talk, I explore how developing humanistic computer vision methods allows us to pursue critical, creative spatial analysis with large map data. Connecting maps as collections of data is a powerful renegotiation of the way the public–not just researchers–can interact with these heritage documents. Using examples from two of my projects (Living with Machines and Machines Reading Maps), I show how working with maps at scale can transform their role in multidisciplinary scholarship.

Bio: Katherine McDonough completed a PhD in History at Stanford University in 2013. Before joining the Living with Machines project at The Alan Turing Institute in March 2019, she taught or was a researcher on digital projects at Stanford, Bates College, and Western Sydney University. She works at the intersection of spatial and textual analysis with colleagues from many disciplines and GLAM curators to develop new methods for creating data from and analyzing multilingual texts and historical maps. At the Turing, she founded the Computer Vision for Digital Heritage Special Interest Group and is the UK PI on Machines Reading Maps.

RSVP: https://pitt.co1.qualtrics.com/jfe/form/SV_bswmYIMeicwgwWq