The Digital Humanities Expert

I’m an assistant professor in the Information School at the University of Washington (UW). My research is focused broadly around artificial intelligence (AI) applications and new ways of searching digital collections held by libraries, archives, and museums. I think of my work in the context of digital humanities and computing cultural heritage. It’s a very interdisciplinary environment: I do work that is computational in nature while also drawing from the humanities.

Currently, I work with both digitized and born-digital archives, such as the End-of-Term Web Archive. Every four years there’s an effort to crawl certain .gov and .mil domains—such as federal agency websites—to preserve their content for posterity. This results in massive web archives that contain hundreds of millions of PDFs. I’m excited about how we can go beyond basic keyword searches and reimagine the visual content embedded within these documents.

I majored in astrophysics and mathematics with a minor in physics at Harvard College. In addition to coursework, I did research on telescope survey pipelines, addressing questions such as how we identify stars in crowded images like those of globular clusters, where deblending stars is a challenge. It was through this research that I first became interested in image processing.

In college I took a course on digital humanities, and when I graduated, I spent a year working at the United States Holocaust Memorial Museum on a digital humanities fellowship. My grandmother survived the Auschwitz-Birkenau concentration camp, and spending that year at the Holocaust Museum was deeply meaningful and important for me both personally and professionally. During my time at the museum, I became interested in how we can understand and search archives in new ways. I worked on a project to categorize millions of documents related to victims of Nazism in the International Tracing Service Archive—an archive that also contains records for my family—using computer vision. It was this project that led me to the computational techniques that I continue to use to this day.

After my fellowship at the Holocaust Museum, I was excited to explore these methods further and started my PhD in computer science at UW. I was fortunate to be an innovator-in-residence at the Library of Congress while working on my dissertation, and I created a project called Newspaper Navigator. Here, I applied a lot of the ideas that I was working on at the Holocaust Museum to a collection of 16 million newspaper pages called Chronicling America. The goal was to use machine learning to extract all of the visual culture embedded within these pages, from photographs to maps, and reimagine how people can search the images visually. After defending my PhD, I returned to the Library of Congress for a yearlong fellowship with the Kluge Center. From there I started as a faculty member at UW’s Information School.

It’s a really exciting time to be in digital humanities. In addition to the long tradition of existing approaches, new methods with AI are changing how we understand our collective cultural heritage. In combining technical and humanistic standpoints, digital humanities is also a perfect place to examine the ethics of AI and datafication. I’m encouraged to see a lot of genuine interest in digital humanities, with so many people doing exciting work in the field.

Physics is a fantastic starting point for thinking about and exploring new ideas and new careers. I actually took no computer science courses during undergrad. My astronomy and physics research and coursework are what led me to the path that I’m on today. Having a background in physics and math has enabled me to grasp coding and machine learning and has paid off enormously. Although the telescope images that I worked on as an undergraduate are ostensibly very different from digitized newspapers or government documents, I’m using what I learned in my time as a physics student every day. Working with telescope survey pipelines—understanding how we handle large volumes of images, how we process them and extract usable information—is not much different from trying to extract useful information from large volumes of archival images in order to make them discoverable.

When it comes to career possibilities, I believe that both curiosity and a willingness to consider different paths can open new doors. Having an interest in history or computer science doesn’t mean that physics isn’t useful anymore—you can use a little of everything. And it’s not just me who feels this way. I’ve worked with mentors and colleagues across all stages of my career who have seen interdisciplinary thinking and collaboration not just as something to explore, but also as something to embrace. My physics background has allowed me to see things differently, and I’m forever grateful for this!