Causally-Informed Reinforcement Learning for Adaptive Emotion-Aware Social Media Recommendation
This addresses the issue of negative emotional impacts from social media recommendations for users, representing an incremental advance by integrating emotion prediction with existing methods.
The paper tackled the problem of social media recommendation systems ignoring users' emotional states by proposing an Emotion-aware Social Media Recommendation (ESMR) framework, which improved emotional recovery and reduced volatility while retaining engagement over 30-day interaction traces.
Social media recommendation systems play a central role in shaping users' emotional experiences. However, most systems are optimized solely for engagement metrics, such as click rate, viewing time, or scrolling, without accounting for users' emotional states. Repeated exposure to emotionally charged content has been shown to negatively affect users' emotional well-being over time. We propose an Emotion-aware Social Media Recommendation (ESMR) framework that personalizes content based on users' evolving emotional trajectories. ESMR integrates a Transformer-based emotion predictor with a hybrid recommendation policy: a LightGBM model for engagement during stable periods and a reinforcement learning agent with causally informed rewards when negative emotional states persist. Through behaviorally grounded evaluation over 30-day interaction traces, ESMR demonstrates improved emotional recovery, reduced volatility, and strong engagement retention. ESMR offers a path toward emotionally aware recommendations without compromising engagement performance.