It has long been observed that that the complexity of computer desktop applications has been growing rapidly, accommodating continuously emerging user needs. In today's desktop environments, users have to assess large amounts of information and complete tedious and repetitive actions. Adaptive user interfaces have been examined as an alternative for reducing the cognitive and motor demands of such tasks. Adaptive user interfaces exploit information about past usage patterns to infer user goals and facilitate tasks. Although the existence of usage patterns has long been verified, previous experimental studies have questioned the effectiveness of adaptive user interfaces. Moreover, the introduction of adaptation techniques in commercial applications has had limited success.
I will discuss the shortcomings of adaptive user interfaces. I will claim that an in-depth understanding of the pros and cons of adaptive user interfaces requires the systematic experimental assessment of their costs and benefits, showing that such costs and benefits should be evaluated with respect to the accuracy of the underlying inference mechanisms. The work that I will present suggests that adaptations should be initiated by the actual users based on their perceived needs for assistance and be supported by visual cues and continuous feedback. I will discuss how this approach can be applied to facilitate the acquisition of targets in desktop environments and demonstrate its use to the access of drop-down cascading menus. This work is part of my Ph.D. thesis at the University of Toronto and has been conducted under the supervision of m.c. schraefel.
Theophanis (Fanis) Tsandilas is a Ph.D. candidate in the Computer Science Department at the University of Toronto. He received his M.Sc. degree in Computer Science at the University of Toronto and his B.Sc. degree in Electrical and Computer Engineering at the National Technical University of Athens. His background is in Artificial Intelligence and his research interests are in Human-Computer Interaction. His dissertation work focuses on the evaluation and design of adaptive and adaptable user interfaces.