The FED-tWIN research profile BALaTAI is a collaboration between the Royal Institute for Cultural Heritage in Brussels and the Research Group for Artificial Intelligence and Sparse Modelling (GAIM) of Ghent University. The goal of the research is to valorize KIK-IRPA’s digital collection and to raise new research questions with innovative artificial intelligence-technologies. These technologies will be applied to the digital art-historical data stored in BALaT (http://balat.kikirpa.be). The FED-tWIN resarch will focus on:

● developing a deep active learning (DAL) framework for detecting objects and features of interest.

● designing an innovative content retrieval and structure learning (coined here CRSL) framework for unsupervised identification of related works (to find visual links or clusters).

● developing a knowledge discovery graph and semantic search principles.

● building a generous interface that shows the richness of the collection and encourages exploratory browsing.

● integrating a visual storytelling platform with interactive functionalities.

These technologies and tools hold the potential of raising new research questions, optimizing human-language queries and discovery of content, increasing audience engagement and attracting new visitors to BALaT.