'First-Hand Accounts of War': identifying names in war letters
In 'First-Hand Accounts of War', CLARIAH Work Package 6 collaborates with the NIOD Institute for War, Holocaust, and Genocide Studies to digitize war letters from the period 1935-1950 into an enriched, searchable dataset.
In the project 'First-Hand Accounts of War: War Letters (1935-1950) from NIOD digitized' more than 150,000 pages of personal correspondence are digitized. The collection includes correspondence, postcards and airmail letters from the period before, during and after the German occupation of the Netherlands and the decolonization war in Indonesia. 'War from First Hand' goes beyond just scanning the material by also developing an enriched and digitally searchable dataset.
Based on the archival descriptions, we know who wrote many of the letters in the collection. Often we also know to whom they were addressed. In this Clariah use case we delved deeper into the material: we tried to automatically map names of people mentioned in the letter texts. We also wanted to do this for names of countries, places, and organizations. In collaboration with Julia Neugarten and Joris van Zundert we developed a work plan using the so-called Named Entity Recognition (NER). With this, we can now largely automatically distinguish and extract names of people, places and organizations from correspondences.
Structuring this information in the digitized war letters makes it easier to digitally search sources, find them and connect them to other collections. In addition, it offers valuable opportunities for a more overarching perspective in historical research on, for example, the thinking of and about different individuals.
'First-Hand Accounts of War' is made possible by funding from the Mondriaan Fund, the Ministry of Health, Welfare and Sport and the NIOD Institute for War, Holocaust and Genocide Studies.
Project info
Researchers
WP6 Co-leader, CLARIAH NL
Employee until October 2022, KNAW Humanities Cluster