情報処理学会 第86回全国大会 会期:2024年3月15日~17日

4M-03
EATOUT: Leveraging the Dynamics of Leadership in Group Recommendation Systems
○兪 佩錦,木實新一(九大)
In the domain of group recommendation systems (GRS), addressing the preferences of individuals with similar tastes is of utmost significance. Given the inherent diversity of opinions within a group, existing GRS techniques typically aggregate individual preferences into a collective group preference for providing recommendations. This paper introduces an innovative approach to group recommendation systems, with a specific focus on small groups sharing common interests. We introduce a web-based restaurant recommendation system designed for increased user satisfaction within small-scale groups. Drawing insights from group decision-making literature, we propose a recommendation algorithm centered around leadership. The proposed algorithm is designed to foster consensus among group members, leveraging the dynamics of leadership, and ultimately increasing the overall satisfaction of the group with the recommended choices.