Data Science of Digital History

Widespread digitization of historical artifacts and the explosive growth of online sources has catapulted historians’ work from an “age of scarcity” to an “age of abundance.” Where historians once had to spend months or years painstakingly collecting documents, today they can find or photograph thousands in hours. Unfortunately, historians’ traditional methods were never developed to cope with this bounty. Making sense of such “big data of the past” requires new approaches to data, analysis, visualization, and interpretation – a project that must draw on expertise from the disciplines of both history and data science.

To assemble and share this expertise, we propose to create a new doctoral training unit, “Deep Data Science of Digital History” (D4H) which will train a new generation of PhD students who work confidently across scientific disciplines; D4H will encourage collaboration across faculties and fields by exploring deep history and deep data science.

D4H organizes its research into three thematic and two functional categories. The thematic categories tackle three key intersections of history and data science:
1. Deep data & knowledge explores the unique challenge of creating historical datasets. Unlike in most other scientific disciplines, historians are comfortable working with different types of sources which each require contextualization
in order to understand their meaning. This research axis of D4H focuses on the characteristics, formats, histories, and infrastructures of historical data in order to train PhD students in historical data criticism and traceable data management.
2. Deep analytics & learning engages with state-of-the-art approaches in machine learning technologies and the use of artificial intelligence for analyzing large historical datasets. This D4H axis evaluates the potential of statistical modelling and and simulations for developing historical questions at scale.
3. Deep visualization & interpretation explores how visualization techniques and user interfaces transform historical imagination and interpretation. This axis of D4H will spark debates about how historical arguments can be turned into
graphic arguments, and how new techniques of representing big historical datasets can be used to explore historical information across time and space.

Cutting across these three themes, D4H will also investigate two functional categories:
1. Digital history and hermeneutics builds on the successes of an earlier digital history doctoral training unit and carries those lessons forward to lay D4H’s foundation;
2. Deep time & history provides a space for contemplating how the collection, analysis, and interpretation of the big data of the past transforms our understanding of time.

We have designed all of our categories to function as friendly “trading zones,” where faculty and students alike can build fruitful bridges among disciplines and expertise to develop a new shared understanding of historical research.