Digital History Workshop – Text-Mining

The Digital History Workshop is a monthly gathering during which researchers in the humanities can become familiar with digital methods and tools. No prior knowledge or skills are expected. Everyone is welcome!

In the upcoming Digital History Workshop, we explore how text-mining helps us understand the history of public health in interwar London.

Description: Can computational methods explain how context shapes (or is reflected in) text? To answer this question, we will scrutinise the “London’s Pulse” corpus, a collection of Medical Officer of Health reports that provide comprehensive information about the general state of health in the boroughs of London. Using Topic Modelling (more specifically Structural Topic Models) we analyse how various themes in these reports are associated with specific external factors, such as the level of poverty in a borough or the political orientation of the local council. In short, this workshop provides researchers with computational techniques to understand the relation between a content and context.

Registration: Please register by sending an email to k.beelen@uva.nl

Software: RStudio (https://www.rstudio.com/)

Data: London’s Pulse raw text corpus (http://wellcomelibrary.org/moh/)

Kaspar Beelen works as a researcher at CREATE. His research focusses on computational history, more specifically on text-mining historical corpora.

The workshop will take place on 20 June, 15.00h-17.00h, Turfdraagsterpad 9, BG 1, Room 0.13.