Corpus Palaeography: Machine Learning, Scribal Profiling and the Dating and Localisation of Manuscripts Containing Old English, c. 800–1200
- Event type
- Seminar
- Series
- Medieval Manuscripts Seminar
- Address
- Hybrid via Zoom, and in the Dr Seng T Lee Centre for Manuscript and Book Studies, Senate House Library, Malet Street, London WC1E 7HU
- Speakers
- Mark Faulkner (Trinity College Dublin), Elisabetta Magnanti (Trinity College Dublin)
- Event dates
- Contact
- [email protected] Email only
Recent innovations in machine learning and digital typesetting offer the scope for a paradigm shift in philological data extraction, analysis and argumentation, where texts are compared not on the basis of generalisation and exemplification, but millions of individual datapoints. Through an Handwritten Text Recognition (HTR) model, trained on c. 800 pages (c. 250,000 words) of Old English to recognise a character inventory of almost 600 letter-forms and marks of punctuation with a character error rate of just 4.15%, we show the potential for a new corpus palaeography.
Unless stated otherwise, all our events are free of charge and anyone interested in the topic is welcome to attend. Registration is required for all events.
