Machine Learning in the Support of Context-Aware Adaptation

Genaro Motti, Vivian [UCL] Mezhoudi, Nesrine [UCL] Vanderdonckt, Jean [UCL]

Adapting user interfaces according to the context of use aims at improving the usability levels of an application and enhancing the user experience, mainly by optimizing the users’ interaction and reducing their errors. However, given the significant amount of information involved, adapting UIs often demands complex inferences. Because the context information is extensive, it is hard to prioritize it to decide the best adaptation techniques. Moreover, dealing with recurrent trade-offs, e.g. adaptability vs. performance, is not simple. To aid the adaptation decisions, machine learning algorithms can be applied to support reasoning, inferences and also to deal with complex or fuzzy information. Although ML can provide several benefits for CAA, there is no agreed framework that aids developers in applying it. Thus, aiming to fill such a gap, this paper defines potential scenarios of CAA where ML can be successfully applied, presenting their common requirements and main trade-offs.