Using Machine Learning to Generate Big Data for Book History
Tuesday, 6 December 2022, 10 – 11:30am 
A seminar by Mark Faulkner (School of English, Trinity College Dublin) and Elisabetta Magnanti (University of Vienna) as part of the Manuscript, Book and Print Seminar Series, organised by the Manuscript, Book and Print Research Theme and the Library of Trinity College Dublin.
This paper showcases the affordances of Handwritten Text Recognition, based on machine learning, for rapidly creating transcriptions of large corpora of medieval manuscripts that can then be analysed en masse. In particular, it showcases work towards version 2 of TOXIIC (Trinity Old English from the XIIth Century), and discusses some of the epistemological and methodological consequences for textual philology of having ready access to such materials.
This new seminar series showcases the diversity of research conducted in Trinity related to the research theme in Manuscript, Book and Print Cultures, which among many other topics includes a focus on historical questions around the origins and development of writing systems, literatures and literary cultures, with research on issues related to textuality in society like literacy, reading and wellbeing, and policy issues around maximising access to Ireland’s written cultural heritage. Papers address engaging children in reading, the wellbeing benefits to reading in Old Age, the experience of working as a creative writer with illustrators, manuscript culture, and digitisation and the new possibilities it opens up for research in this field.
Campus Location: Trinity Long Room Hub Arts and Humanities Research Institute
Accessibility: Yes
Room: Neill Lecture Theatre
Research Theme: Manuscript, Book and Print Culture
Event Category: Arts and Culture, Lectures and Seminars, Library
Type of Event: One-time event
Audience: Researchers, Faculty & Staff, Public
Cost: Free
Contact Name: Dr Mark Faulkner
