The complex nature of intelligent systems motivates work on supporting users during interaction, for example, through explanations. However, as of yet, there is little empirical evidence in regard to specific problems users face when applying such systems in everyday situations. This article contributes a novel method and analysis to investigate such problems as reported by users: We analysed 45,448 reviews of four apps on the Google Play Store (Facebook, Netflix, Google Maps, and Google Assistant) with sentiment analysis and topic modelling to reveal problems during interaction that can be attributed to the apps’ algorithmic decision-making. We enriched this data with users’ coping and support strategies through a follow-up online survey (N = 286). In particular, we found problems and strategies related to content, algorithm, user choice, and feedback. We discuss corresponding implications for designing user support, highlighting the importance of user control and explanations of output rather than processes.

The complex nature of intelligent systems motivates work on supporting users during interaction, for example through explanations. However, there is yet little empirical evidence on specific problems users face in such systems in everyday use. This paper investigates such problems as reported by users: We analysed 35,448 reviews of three apps on the Google Play Store (Facebook, Netflix and Google Maps) with sentiment analysis and topic modelling to reveal problems during interaction that can be attributed to the apps' algorithmic decision-making. We enriched this data with users' coping and support strategies through a follow-up online survey (N=286). In particular, we found problems and strategies related to content, algorithm, user choice, and feedback. We discuss corresponding implications for designing user support, highlighting the importance of user control and explanations of output, not processes. Our work thus contributes empirical evidence to facilitate understanding of users' everyday problems with intelligent systems.

With recent progress in display technology, visual see-through head-mounted displays are beginning to enter our everyday lives. Especially in cars they may replace head-up displays, as they can theoretically perfectly imitate them but are more flexible to use. However, prior work has shown that both screen- and vehicle-stabilized content suffer from drawbacks such as occlusion or technological limitations. As a potential alternative, we propose three concept alternatives, in which head rotation is used to manipulate the displayed content differently from both of the known stabilization techniques. In a qualitative user study, we identify the best concept proposal and then evaluate it against the established content stabilization techniques. The presented concept is perceived to be more applicable for the proposed use case and effectively reduces some of the known problems of both stabilization techniques.