Navigating the Challenges of Deduplicating Actors in Historical Letter Exchanges




Actor deduplication, Historical Correspondence, Epistolary Data, Semantic Web, Knowledge Graphs


This paper outlines ongoing research to standardize the metadata of letters from Finnish cultural heritage organizations to build a comprehensive archive for studying 19th-century epistolary culture in the Grand Duchy of Finland (1808/09-1917), as a part of the Constellations of Correspondence (CoCo) project. The challenge of deduplicating and reconciling actors and their associated information is an important problem faced in historical knowledge databases. The actors, e.g., senders and recipients of letters, can be individual people, couples, or families as well as organizations like institutions or companies. The deduplication described in this article has its focus on the individual people. The solutions we propose in this paper are significant for the development of the efficient and accurate approaches in this field.