User Interface and Architecture Adaption Based on Emotions and Behaviors

This paper shows how emotions and behavior considerations in socio-technical systems lead to high-quality self-adaptations, both at application and architecture levels. In our approach, an interactive control system assesses the reconfigurations that enhance the quality of service (QoS) while considering humans’ quality of experience (QoE). We use a Model-Free Reinforcement Learning (MFRL) approach to self-adapt user interfaces (UIs) to users’ emotions. The approach aims to maximize applying the essential adaptations and minimize the unnecessary ones towards users’ QoE, i.e., task completion and satisfaction. If the control system detects a drop in QoS in emotion-based adaptations or other functions, another level of adaptation reconfigures the architecture towards better quality. We chose emergency evacuation training as a suitable evaluation domain since people experience intense emotions in such potential danger. We performed experiments with a mobile application we developed that acts as a recommender system in emergency training. By taking contextual input of the users’ basic emotions from face recognition and their mobility behavior, the application intelligently adapts its UI to quickly lead people to safe areas while keeping them emotionally stable. In addition to UI adaptation, the system is capable of architecture-level adaptations to decrease response time if required. The evaluation process confirms the efficiency of the MFRL in iterations, as well as compared to other possible UI adaptation techniques. The emerging results also show that architecture-level adaptations positively impact the system performance and users’ emotions and performance.