Supporting Adaptation Intelligibility, Control, Evaluation, and Adaptation Policy Evolution in a Context-based Adaptive Collaboration Environment
Syed Sajid Hussain
erschienen in:

Nowadays knowledge-intensive collaborative work is performed by distributed teams of end-users. Collaboration Environments (CEs) enable such collaborative work by providing a variety of affordances, i.e., through applications, processes and artifacts. These affordances need to be continuously adjusted in order to support the needs of end-users in emergent collaboration situations. A context-based adaptive shared workspace CE automatically adjusts its affordances based on context representation and adaptation knowledge. Such an automatic adaptation poses challenges for the end-users: they may not notice an adaptation, or may not be able to understand the cause and effect of the adaptation. Also, since no system is perfect, an adaptation may not fit the needs of the end-users in a collaboration situation. Therefore, an end-user affected by the adaptation may feel the need to make an exception from the adaptation. Another affected end-user may be happy with the original adaptation and may not understand the exception. This may lead to a social conflict among the affected end-users. Since an Adaptation Policy translates the team’s agreed-upon collaboration strategies into the adaptation of the CE, the end-users can best evaluate the quality of the adaptation in-situ. Moreover, the end-users, after evaluating the adaptation, may feel the need to modify/delete an erroneous Adaptation Policy or to define a new Adaptation Policy. This thesis takes the end-users’ viewpoint on context-based adaptation of a shared workspace CE. It identifies four research problems to be solved. The research problems are adaptation intelligibility, end-users’ control over adaptation, adaptation evaluation, and Adaptation Policy evolution. This thesis proposes the sensemaking process model, which guides the end-users’ activities of making sense of the adaptation by solving the above research problems. A context-based adaptive shared workspace CE is extended by the application support functions for providing the execution support for the sensemaking process. The results of two pluralistic usability walkthrough experiments show that the end-users can use the application support functions and consider them beneficial. In this thesis it is shown that the end-users can evaluate and modify the adaptive behavior of the shared workspace CE. Therefore, this thesis enables future research to evaluate the usage and efficiency of adaptations.

Dirk Veiel | 12.08.2021