Metadata is data about data. It plays a key role in allowing people to access and use the information in libraries and archive catalogues. Despite recent technical advances of AI for interpreting text and images, the generation of metadata for large collections of historical books, images, and texts remains an unsolved problem. In this project, we work together with the KB National Library of the Netherlands in order to put explainable AI to use in libraries. For example, the KB National Library is seeking ways to enrich metadata for illustrations contained in important portions of its historical book collection and the library is also in need of technology that can establish new links between books in their million-scale collection that explain how the books are related. In this multidisciplinary PhD project supervised by Dr Iris Hendrickx and Prof. Martha Larson, you will combine insights from explainable AI, data science, information retrieval, natural language processing and library science to address the practical challenges faced by large libraries.
Plus d’informations :
[Website Radboud University]