NWO, the Netherlands Organisation for Scientific Research awarded funding from the CATCH (Continuous Access to Cultural Heritage) thematic programme to the new Hitime (Historical Timeline Mining and Extraction). HITIME aims to develop a text mining toolkit for historical texts. The project addresses major challenges in the area of text mining, such as the language-independent detection of significant events, time expressions, historical spelling variations and the ability to cope with stylistic diversity.
HITIME will generate a "Web of History Knowledge" on the basis of a growing corpus of historical source material. This associative network of concepts (events, persons, organizations, occupations, time expressions, and geographical locations) allows one to find and visualize large-scale sets of related sources, using interactive timelines and maps. The toolkit will offer the means to answer new research questions (such as on social networks, group biographies) and to semantically enrich archival aids and other text sources.
Partners in the project are: TiCC (Tilburg centre for Creative Computer, Tilburg University), IISH (International Institute of Social History), Institute of History at the University of Leiden and the VKS (Virtual Knowledge Studio). The project leader is prof. Antal van den Bosch (TiCC), and the project duration is four years.