Emotional Internet of Behaviors: A QoE-QoS Adjustment Mechanism

The Internet of Behaviors (IoB) approach supports developing socio-technical systems based on humans’ goals, characteristics, behaviors, and emotions. This paper shows how emotions and behaviors could impact the quality of software systems. We propose interactive control loops that supervise application and architecture adaptations toward enhancing the system quality of service (QoS) and human quality of experience (QoE). Under the IoB conceptual model, we first show how historical and real emotions could be the source of the design and adaptation of socio-technical systems. We further use a Reinforcement Learning (RL)-based approach as a self-adaptation supervisor of user interfaces (UIs) to users’ emotions. The approach aims to maximize applying the essential adaptations and minimize the unnecessary ones towards users’ QoE. 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 used the emotional IoB approach to develop a mobile application as a recommender system in emergency evacuation training. The app takes users’ facial emotions and positions as input and adapts its UI to impact users’ target emotions and task completion. In addition to UI adaptation, the system supports architecture adaptations to decrease response time if required. The evaluation process confirms the efficiency of the RL in iterations, as well as compared to other possible UI adaptation techniques. The results also show that architecture adaptations positively impact the system performance and users’ emotions and performance.