Usability Enhancing Services

ContactMichael Quade, Mathias Runge, Maximilian Kern


The research field "usability enhancing services” deals with methods and techniques for raising the usability of novel, adaptive user interfaces. A majority of this science is focusing on the formalization of user attributes (properties, desires, etc.) and their impact on system interactions. Based on this formulization user models are created, which are utilized by methods of automated usability evaluation, as well as adaptive assistance.

Automated Usability Evaluation

DAI-Labor deals with the research of future user interfaces. By now, it is obviously that these interfaces became multi-modal and adaptable to the situation of the user, his preferences and the system itself. This leads to completely new challenges for usability evaluators. On the one hand, they have to evaluate all variants of an adaptive user interface and on the other hand, the underlying system information is not completely known at design time. An optimal evaluation of adaptive user interfaces take place during design time and runtime as well.

At design time, a user model supports the designer by automatically interacting with the system or a model of the system the designer want’s to create for early usability problems discovery. Above that, numerous variants of the interface can be rapidly evaluated. Dependencies for the adaptation will be emulated but the complete data is only available at the concrete runtime. Therefore, a user model will be also part of the system and evaluates the adaptations of the user interface in relation to the context of use, the user and the system.

Adaptive assistance

The capabilities of devices are rising continuously. As a consequence, the interaction with devices becomes more complex. Adaptive software systems antagonize this complexity by adapting to the current context-of-use and only considering available interaction resources within the user interface. Thereby, the resulting appearance of the user interface might become hard to predict through the application of multiple adaptations. Because of that, the system information is utilized during run time in order to make use of the actual state of a system. Such as highly adaptive software system requires a help functionality which is also adaptive.

The information shown within a user interface reflects only a small portion of the information content of the available system information. The neglected information is utilized at run time in order to generate automatically assistance. Assistance is requested either explicitly by the user or pro-actively provided from the system by observing her behavior.