Research Proposal: Using Computational Design to Enhance Emotion-Driven Recommendations in Multimedia Experiences

This position paper proposes a novel computational design approach for enhancing emotion-driven recommendations in immersive multimedia experiences. This approach aims to address the limitations of current recommendation systems that often lack the capability to understand and cater to users’ emotional states. As multimedia experiences have become an integral part of our daily lives, platforms increasingly rely on personalized recommendations to engage users. An optimized user quality of experience (QoE) is crucial. The goal in this research is to explore the incorporation of emotional aspects, specifically using physiological measures, in designing and developing improved recommender systems. We focus on the use of computational design methods to create adaptive recommender systems, examining deep reinforcement learning as a potential approach. Specifically, we explore the combination of recurrent neural networks and reinforcement learning to create a proof of concept. This paper aims to provide insights into the justification for utilizing computational design in the development of emotion-driven immersive multimedia recommendation systems.