In “Hospital@Night”, the need for more efficient task scheduling for on-call physicians in after hour clinical settings is addressed through state-of-the-art machine learning and visual analytics methods. Its idea is to improve the often arcane and complex manual scheduling practices through phone and pager calls currently employed in clinical task distribution. More often than not, the inflexibility and heterogeneity of current practices overburden and interrupt on-call physicians with requests that do not match their competencies, their current location, or simply their need for personal down time to perform at their best in the high-stakes, high-stress clinical scenarios to which they are called. In Hospital@Night, health IT specialists from Systematic, computer science researchers from AU, and clinicians from the University Hospitals in Aarhus and Aalborg set out to change this status quo through data-driven methods that automatically distills situational and domain knowledge on clinical task scheduling in a knowledge graph, which can then be used as a basis for improved task distribution. The role of the research group on Ubiquitous Computing and Interaction lies in the investigation of visualization approaches for this knowledge graph to enable human feedback and modification to the automated process, so as to interactively prune and adjust the complex, generic knowledge graph to the specifics of a particular hospital or ward.