Seminars and talks
Digital Humanities Colloquium 2018
12 December 2018
Room 2.19, Samuel Alexander Building
This year’s event is entitled ‘Digital Approaches to Ancient and Modern Texts’. We hope that the event will attract an audience not only of classicists and ancient historians, but anticipate that by showcasing classicists' deployment of digital approaches we are also able to initiate conversations with those who take digitally-informed approaches to medieval and modern texts.
This event showcases a number of digital approaches to the interpretation, decipherment and analysis of ancient and modern texts, in particular those which are extant in material form. It will begin with a keynote from Charlotte Tupman, Research Fellow in Digital Humanities at the University of Exeter. We hope that the event will attract an audience of classicists and ancient historians, but we anticipate that by showcasing classicists' deployment of digital approaches we would also be able to initiate conversations with those who take digitally-informed approaches to medieval and modern texts.
- Dr Charlotte Tupman (University of Exeter): Towards a world of historical linked open data: where can our texts take us?
- Dr Lise Jaillant (University of Loughborough): Textual Futures, Archives and Small Poetry Publishers: the Carcanet Press
- Prof. Klaas Bentein (University of Gent): Variation equals social meaning? Developing a digital approach towards communicative variation in Roman and Late Antique Egypt
- Prof. Rodney Ast (University of Heidelberg): What the Digital’s Been Doing with the Humanities: The Case of Papyrology
- Dr Henry Jones (University of Manchester): Using Digital Corpus Analysis Tools to Study Successive Retranslations of Thucydides' History of the Peloponnesian War: Insights from the Genealogies of Knowledge project
- Prof. Stephen Todd (University of Manchester): Classical Philology vis-à-vis Corpus Linguistics: Some methodological diffractions
Match, Compare, Classify and Annotate: Computer Vision for the Digital Humanities
Dr Giles Bergel, University of Oxford
Monday 4 June 2018
Room A1.12, Samuel Alexander Building
Computer vision has made rapid progress in recent years: computers can now reliably match the same image; find differences in similar images; and classify content within multiple images. Recently (and controversially) computers have also begun to be able to identify subjects, such as people and places. This hands-on session will show how to install and use state-of-the-art open source software from the University of Oxford’s Visual Geometry Group.
Attendees will leave the session with the knowledge of how to match, differentiate, classify and annotate content within many kinds of images. The event will also include a short presentation of Dr Guyda Armstrong's Envisioning Dante project, a partnership between the John Rylands Research Institute and the Oxford Seebibyte Research Group which is using these tools to classify and analyse the page design of early printed editions of Dante's Divine Comedy held in the Rylands Special Collections.
Requirements: a laptop or tablet is needed to access online demos. Attendees who would like to try the software on their own laptops will need to install Docker, ideally in advance of the session (see links to the various platforms under Docker Version. There is no need to install the tools themselves or the training data as this will be supplied on the day).
In addition, Giles will be available for short consultations with those wishing to discuss their research cases.
This workshop is offered in association with The John Rylands Research Institute.
The State of Stylometry: Achievements and Challenges in Computational Stylistics
Prof Dr Mike Kestemont, University of Antwerp
Thursday 26 January 2017
Room A4, Samuel Alexander Building
Stylometry or computational stylistics is a popular subfield of the Digital Humanities, an international community of scholars which explores how computational methods can enhance the existing practice in Humanities scholarship.
Computational authorship studies have probably been stylometry's most successful application recently, with high-visibility case studies, involving J. K. Rowling or Harper Lee, attracting much attention in the popular media.
In this paper, we will introduce and survey the state of the art in stylometry, with ample attention for some of the most difficult challenges which remain, such as cross-genre authorship attribution and authorship verification.
Throughout the talk, we will refer to representative work in recent stylometry, including a recent authentication study of the War Commentaries of Julius Caesar, which sheds a fascinating light on the oeuvre traditionally attributed to the Roman general.
Mike Kestemont is a research professor in Digital Text Analysis at the University of Antwerp. He specializes in computational text analysis for the humanities, in particular stylometry or computational stylistics. He has published on the topic of authorship attribution in various fields, such as classics or medieval European literature.
Mike actively engages in the debate surrounding the digital humanities and attempts to merge methods from artificial intelligence with traditional scholarship in the humanities. His website contains pointers to his recent scholarly activities, including an open access scientific documentary about stylometry and Hildegard of Bingen.