Sequential Group Recommendations with Responsibility Constraints

Abstract

This paper addresses the aspects of fairness and transparency in Group Recommender Systems (GRS), crucial for fostering user trust and system reliability. With Recommender Systems (RS) being pivotal in modern applications, ensuring responsible recommendation techniques is paramount. We focus on two key responsibility constraints: fairness and transparency. Fairness in group recommendations poses challenges, particularly in the context of sequential recommendation rounds. We introduce the concept of sequential group recommendations, emphasizing the need to consider multiple rounds to prevent bias against any group member and ensure overall satisfaction. Additionally, we explore transparency in recommendations, proposing explanations for why certain items are excluded from recommendation lists. Our work enhances the ethical and user-centric aspects of web engineering, which are essential for the success of web applications.