Designing easy-to-use user interfaces is a challenging problem-solving task. The number of possible designs in any problem is too large to explore manually. For example, there are (2n)! possible ways to organize n menu items to a hierarchical menu system. For just 50 items, there are more than 10^157 alternative designs. Moreover, the relevant human factors can be complex. Ideally, the design of well-understood parts of an interface could be automatized, allowing the designer to focus on the truly novel aspects of the problem. This goal can be achieved with predictive models of human-computer interaction combined with either computational optimization methods or interactive visualization techniques.
Whereas previous work in user interface design has been largely based on trial and error, this approach allows aggressive exploration of the design space. The outcomes are demonstrated as novel user interfaces targeted to both classic interactive tasks, such as text entry and menu selection, as well as novel multimodal and mixed reality interfaces.
Affiliation
The group is part of M2CI, the German Cluster of Excellence in Multimodal Computing and Interaction. The cluster was established by the German Research Foundation (DFG) within the framework of the German Excellence Initiative in November 2007.
At MPI-INF, the group is affiliated with Department 4: Computer Graphics. The group is also affiliated with the Max Planck Center for Visual Computing and Communication (MPC-VCC) that was established by the Max Planck Society for the Advancement of Science (MPG) and Stanford University in October 2003. The Max Planck Center is funded by the Federal Ministry of Education and Research of the Federal Republic of Germany (FKZ 01IMC01, FKZ 01IM10001).