About the project
One major challenge in history research is tracking ordinary individuals across different settings. An individual might be well documented in multiple sources, but the masses of text in question are too large for any systematic approach.
This project aims to build a workflow to assist in that challenge. The project uses natural language processing (machine learning) to extract key information from Danish eighteenth-century text. This is done by manually annotating data to train a model that automatically identifies similar information in other texts.
The categories extracted include names, occupations and verbs. The project will also experiment with annotating phrases that describe bodies and faces.
Participants
Johan Heinsen, Department of Politics and Society (Faculty of Social Sciences), heinsen@dps.aau.dk
Kristian Gade Kjelmann, CALDISS (Faculty of Social Sciences), kgk@adm.aau.dk
Anders Dyrborg Birkemose, Department of Politics and Society (Faculty of Social Sciences)
Assistants
Armin Pasalic, student, apasal19@student.aau.dk
Nana Ohmeyer, student, nohmey18@student.aau.dk
