CREATE Salon: Text Mining (13 December)

The upcoming CREATE Salon takes place on Thursday 13 December between 3:00-5:00pm, eLab Mediastudies (BG1 room 0,16) Turfdraagsterpad 9, 1012 XT Amsterdam. This month’s topic is text mining, with presentations by Lucas van der Deijl and Kaspar von Beelen.

Lucas van der Deijl (UvA): Automatic text comparison and early modern translation
This presentation demonstrates the use of automatic text comparison for the study of early modern cultures of translation. By adopting procedures and metrics from the field of machine translation and automatic text generation, different translations of the same source text can be aligned and compared automatically. Based on various parallel translations from different genres, this method reveals the reuse, revision and dissemination of vernacular texts during the seventeenth century. Finally, this presentation argues for the additional value of digital text analysis, not only on the macro level of the corpus, but also on the micro level of individual texts, sentences and words.

Kaspar Beelen (UvA, CREATE): Mining ‘London’s Pulse’. Computational approaches to the Public Health Discourse of Interwar London
This presentation comprises a computational analysis of the discourse on public health in interwar London. It uses the London’s Pulse corpus, a collection of annual reports written by the Medical Officers of Health (MOsH). These officers were attached to the local authorities in and around London, and wrote extensively on their observations and encounters that resulted from their inspections. While London’s Pulse includes all the available reports from 1848 to 1972 (c. 121,311,432 words) we are primarily concerned with the interwar years, which have been of special interest to historians of public health. Using various computational techniques, the presentation analyses the language concerned with personal prevention. It finds cause to dispute the supposed reorientation of interwar public health work towards encouraging individuals to adopt healthy behaviours and take care of themselves, and also its continued resistance to state-led approaches prior to the creation of the National Health Service in 1948.


Lucas van der Deijl is a PhD candidate at the University of Amsterdam funded by NWO’s ‘PhD’s in the Humanities’. He studies the early modern translation and adaptation of Descartes and Spinoza in Dutch from a quantitative perspective. This research was prepared during a Pre-PhD fellowship at CREATE in 2016/2017.

Kaspar Beelen currently works as Assistant Professor Digital Humanities at the University of Amsterdam. After completing his Ph.D., he worked on the Digging into Linked Parliamentary Data Project at the University of Toronto and the University of Amsterdam. As researcher for CREATE he focusses on computational history, more specifically on the use of text-mining for political and cultural history. His main areas of interest include: gender and politics, the history of political representation and the evolution of affective discourse.