Computational linguistic opportunities at the Koninklijke Bibliotheek

The National Library of the Netherlands (KB) is an active partner in national and international cooperative efforts to develop new knowledge and technology. With this poster, we will showcase what the KB can offer researchers in the field of computational linguistics. We will present our digitised datasets, current research projects and the services of the KB Lab. The Koninklijke Bibliotheek (KB), National Library of the Netherlands, is a research library with a broad collection in the fields of Dutch history, culture and society. As a national library we collect and store all (digital) publications that appear in the Netherlands, as well as a part of the international publications about the Netherlands. The KB has planned to have digitised and OCRed its entire collection of books, periodicals and newspapers from 1470 onward by the year 2030. Over 60 million book-, newspaper- and magazine pages are currently available via the search portal www.delpher.nl. Via DNBL (www.dbnl.nl) high quality digitized works of Dutch literature, linguistics and Cultural History can be accessed. To further improve the usability of our content we aim to have all relevant names of persons, locations and organisations in our digital content reusable as linked (open) data in the near future. For this, a team of researchers at the KB Research Department is developing methods to automatically extract these types of information from the unstructured text of e.g. historical newspaper articles and use it to improve the findability and usability of our digital content. Next to this, most of our datasets (such as historical newspapers, books, periodicals and catalogues) are freely available for research purposes via our Dataservices department and we welcome and encourage experiments and new applications. The virtual KB Lab shows some of such applications and invites researchers to experiment with our data, new technologies and innovative prototypes. The KB also collaborates with academic researchers in research projects or (junior) fellowships to learn from their research in order to improve our services. Example outcomes of research projects using computational linguistic methods are the Frame Generator and Genre Classifier. This poster will present the various datasets that the KB has available for research, the activities we undertake to work together with scholars in research projects, the services that we offer those who wish to work with our material and demos of the tools available in our KB Research Lab.