Textual Geographies uses named entity recognition and geolocation to extract place names from
multilingual
(English, German, Spanish, and Chinese) printed volumes held by the HathiTrust digital library and
to
associate those names with detailed geographic information. The project corpus currently includes
about 10 million volumes published between 1700 and the present day.
The project is under active development and has enjoyed generous funding from the
National Endowment for the Humanities Office of Digital Humanities,
American Council of Learned Societies, and the
University of Notre Dame.
For more information concerning the corpus and research methods, please contact the
project director,
Matthew Wilkens.