Report: A charter repository as textual basis and scientific tool for German Medieval Studies (?)
Yesterday, two team members of DiDip presented the project in the halls of the Center for Information Modelling (ZIM) at Graz. The presentation was the first of four on how digital humanists at the ZIM deal with historical data stemming from the Middle Ages.
While the occasion was primarily tailored to potential interests and needs of German Medievalists at Graz University, our team hinted at digital methods that can be employed when analyzing enormous amounts of historical sources. We demonstrated that Monasterium‘s database mostly includes documents of legal relevance – diplomatic charters, a kind of text type on its own. Digressions on the XML technology stack, Computer Vision (CV), and Natural Language Processing (NLP) methods made visible how computers help in structuring, navigating, and processing overwhelming numbers of documents. For different use cases, we illustrated how Text Recognition and Normalization, Part-of-Speech (POS) Tagging, and Named-Entity Recognition (NER) exploit machine intelligence to reduce manual labor on the side of humanist scholars.
Throughout the event, an increased interest in these novel techniques became apparent. We are convinced that not only scholars of history and diplomatics but also of language-centered research fields can greatly benefit from using them. Within DiDip, we are looking forward to creating a virtual environment in which research tools can be used interdisciplinarily and by scholars of many backgrounds.
Slides are available in German.