The National Endowment for the Humanities has announced that it has awarded a research grant to a USC-UK digital humanities team to develop a workflow that would use advanced machine learning and annotation tools to extract and annotate text on maps across large historic map collections.
Dr. Yao-Yi Chiang, associate professor (research) with the USC Spatial Sciences Institute, and Deborah Holmes-Wong, director of the USC Digital Library, are leaders in extracting text from maps. They will collaborate with Dr. Katherine McDonough, a pioneer in analyzing map features and text for historical research at the Alan Turing Institute in England. The research team will work with one of world’s largest sets of scanned historical maps through heritage partners from the British Library, the National Library of Scotland, the Library of Congress, and the USC Libraries.
In this project, the research team will build automatic methods to normalize text in historical maps as a new kind of data that can be used across the humanities and the heritage sector. By building “Machines Reading Maps,” humanists and heritage professionals can tap into a previously inaccessible body of knowledge that has remained locked in historical maps by making the text on maps searchable and linked to other geospatial data and collections.
Innovations in this project include developing the ability to read map content at scale using tools for text (as opposed to images), integrating place entity linking and image annotation tools to make text on maps meaningful, and improving map discovery and collection histories at cultural institutions.
Chiang’s role is to build the AI technologies that can automatically read text from thousands of scanned historical maps and make that text meaningful.
This project will contribute the growth in the scholarly community working with research questions about the spatial experience of industrialization in 19th-century Great Britain and social change in US cities during the 20th century.
Read more about Chiang’s work in spatial computing at https://spatial-computing.github.io/.