Understand and explain decisions of intelligent systems from the user's point of view
We research and develop data-based, intelligent user interfaces designed to satisfy the requirements and needs of humans. These activities focus on the life cycle of modern applications for personalized systems such as stress recognition, pilot-in-the-loop or recommender systems. To do that we develop user-, domain- and task-models and apply machine learning and data mining methods.
An additional focus involves designing understandable AI decision paths and making intelligent systems more trustworthy. For example, we develop models and user interfaces that convey AI decisions in a transparent and understandable manner.
Using human-centric principles as a foundation, our goal is to find persuasive application solutions and to promote the acceptance of intelligent systems in every field of application possible. In the process, we endeavor to design the interaction between humans and learning systems so that it is natural, intuitive, robust and reliable.