Abstract Summary
Previous research on group recommender systems (GRSs) has shown that group dynamics strongly influence decision-making, yet collaborative filtering (CF)¨Cbased GRSs rarely account for social interactions, largely because suitable analytical tools are lacking. This paper introduces a community resource for studying live groups as they engage with a CF-based recommender system through a domain-independent graphical interface that records interaction signals (such as suggestions, views, and favorites) and integrates them into the recommendation process. A live user study with 72 participants organized into 18 groups demonstrates the system¡¯s effectiveness in capturing and analyzing user interactions. Results show that incorporating interaction awareness enhances group satisfaction and reveals underlying social dynamics, offering new opportunities for adaptive GRSs responsive to real-time user behavior. Source code and dataset available online at this link1.
