Digitalisation does not stop at the humanities and thus becomes increasingly important for the work of historians. Digital methods in the field of handwriting recognition (HTR) and computer-assisted natural language processing (NLP) already offer great potential for the analysis of historical sources. Until now, however, these applications have often only been available to those with extensive knowledge of information science. The international research project ‘The Flow: From Deep Learning to Digital Analysis and their Role in the Humanities. Creating, Evaluating and Critiquing Workflows for Historical Corpora’ addresses this problem. The aim of the project is to develop standardised digital workflows based on existing technologies that make it easier for researchers to work with historical sources digitally. Four sub-projects will contribute to this goal by analysing legal and administrative sources from England, the North European Hanseatic region, Switzerland and Ethiopia. In addition to the FGHO, Bielefeld University and the University of Bern are also participating in the project.

The sub-project carried out at the FGHO Lübeck deals with the minutes of the Hanseatic League meetings – the so-called ‘Hanserezesse’ (Hanseatic Resolutions). The extensive collection of resolutions, which we have from 1358 to 1669, forms one of the most important source corpora on the history and development of the Hanseatic League. The aim of the project is to examine the resolutions with the help of digital applications and thus present the first long-term study on the development of the Hanseatic League.

The use of digital tools will be essential for such an analysis. Applications such as Handwritten Text Recognition (HTR) and Natural Language Processing (NLP) methods such as Named Entity Recognition (NER) can help here. As part of ‘The Flow’, the FGHO is therefore working on the creation of suitable HTR models in order to be able to access the extensive source material of the Hanseatic recesses. For the sources from the 17th century, the results of the citizen science project ‘Read!Hanse.Sources.’ can be used for this purpose.