NWA
The Dutch Research Agenda

HAICu: digital Humanities - Artificial Intelligence - Cultural heritage

In the twenty-first century, information comes from an ever-growing multitude of digital sources, challenging citizens and institutions to weigh each type of information on relevance and reliability. Dutch libraries, archives and museums have faced these challenges for centuries. This is even more the case since their multimodal collections are being digitised into terabytes of information containing books, videos, music, maps, illustrations, photographs and handwritten documents. These datasets are indispensable sources for reliable fact checking and data validation and thus could contribute critically to societal resilience. Yet, current techniques cannot deal with the size, complexity and heterogeneity of these rapidly growing big data collections, making it impossible for stakeholders to effectively tap into them. Current Artificial Intelligence (AI) techniques are uniquely suited to provide a solution, jointly addressing text and other modalities. In HAICu, AI and Digital-Humanities researchers, heritage professionals and engaged citizens will forge scientific breakthroughs in AI to open up, link and analyse large-scale and heterogeneous multimodal digital heritage collections in their dynamic context. The challenges posed by the digital cultural heritage domain provide a powerful opportunity to push the boundaries of AI, while adhering to the latest societal requirements for responsible, explainable methods in the construction of multimodal arguments and narratives. We will facilitate user-assisted generation of fact-based narratives in co-creation with user communities and institutions who depend on relevant and reliable digital information. Our new AI-driven methods and user-engineered tools will help to construct meaningful contexts based on available, massive digital cultural heritage data, weighing both historical and contemporary multimodal information. Methods will be evaluated, with representatives of key stakeholders in various usage scenarios.
HAICu will ultimately present an AI-driven multimodal successor to today’s popular Delpher. This will bring our cultural heritage to life, allowing all of society to access the country’s polyvocal past, present and future.

File number
NWA.1518.22.105
Project lead
dr. A. Weber
Lead organisation
Universiteit Twente
Programme
Onderzoek op Routes door Consortia 2022
Funding instrument
NWA L1
Status
Lopend
Theme
Technology and society
Consortium
Rijksuniversiteit Groningen, NHL Stenden Hogeschool, Universiteit van Amsterdam, Universiteit Twente, Vrije Universiteit Amsterdam, NWO-institutenorganisatie, Hanzehogeschool Groningen, Radboud Universiteit Nijmegen, Universiteit van Tilburg, Hogeschool van Amsterdam, Naturalis Biodiversity Center, Fontys Hogeschool, Universiteit Utrecht, Hogeschool Utrecht, Koninklijke Bibliotheek Nederland, Nederlands Instituut voor Beeld en Geluid, Nationaal Archief, Collectie Overijssel, Rijksdienst voor het Cultureel Erfgoed, Seecr, Novitas, Stichting Bevordering Maatschappelijke Participatie, Het Nieuwe Instituut, RHC Groninger Archieven, Maritiem Museum, Nationaal Museum van Wereldculturen, De Ree Archief-systemen, Spinque, KNAW Humanities Cluster, Common Lab Research Infra. for Arts & Humanities [CLARIAH], Forum Groningen, Vrije Protestantse Radio Omroep, University of Amsterdam Library, Target Field Lab, SURF

Part of routes

We cover these questions

Welke vragen hopen we met dit project te beantwoorden? Deze vragen zijn gegroepeerd in de volgende clustervragen.