Problem definition
When looking at the development of online user interfaces, there is a growing demand for a more user-centered experience; a customizable user interface, however, is not always a solution. Allowing users to modify their own user interface only becomes advantageous once the improved user experience outweighs the user’s investment in customizing the interface. For example, a user will likely find it worthwhile to invest time into customizing an interface which they use frequently (such as an online e-mail interface or portal site) as opposed to customizing an interface which they use occasionally (such as a website to book a vacation). The question that then arises is: can a more user-centered experience be created online without user investment? This encourages looking into adaptive user interfaces; dynamic user interfaces that are scripted to learn from users in order to improve the user experience for individual users.
Pat Langley describes an adaptive user interface (AUI) as a “software artifact that improves its ability to interact with a user by constructing a user model based on partial experiences with that user. [...] Improvement should result from generalization over past experiences and carry over to new user interactions.†In theory, the more that a user interacts with an AUI, the better the user experience should become. For this very reason an AUI “cannot exist in isolation, but rather is designed to interact with a human user. [...] AUIs rely on a precious resource – the user’s time.â€1 Therefore, user interfaces that are used frequently by individuals are an ideal environment for AUIs; the more interactions an interface has with a user, the more accurately it can adapt to match the user’s needs and goals. What does this then mean for user interfaces that users only use occasionally?
References- Langley, P. (1999). User Modeling in Adaptive Interfaces. Seventh International Conference on User Modeling, Springer. [↩]
