TY - GEN
T1 - Evaluating Diversity in Sequential Group Recommendations
AU - Zulfiqar, Haider
AU - Lenzi, Emilia
AU - Stefanidis, Kostas
N1 - Publisher Copyright:
© The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.
PY - 2025
Y1 - 2025
N2 - The increasing complexity of user preferences in group activities, such as movie watching, necessitates more dynamic recommendation systems. SQUIRREL, utilizing Reinforcement Learning, adapts to the changing preferences of group members through sequential group recommendations. This paper introduces an enhancement to SQUIRREL by permitting the re-recommendation of items, reflecting more realistic scenarios where users may revisit previously enjoyed content. We implemented and evaluated three distinct reward functions—focusing on maximizing group satisfaction, balancing satisfaction with diversity, and maximizing diversity—to investigate their impact on recommendation diversity and group satisfaction. Experimental results, using the 20M MovieLens dataset, demonstrate that while diversity-focused rewards enhance recommendation variety, it requires careful balance with user satisfaction to avoid diminishing user engagement.
AB - The increasing complexity of user preferences in group activities, such as movie watching, necessitates more dynamic recommendation systems. SQUIRREL, utilizing Reinforcement Learning, adapts to the changing preferences of group members through sequential group recommendations. This paper introduces an enhancement to SQUIRREL by permitting the re-recommendation of items, reflecting more realistic scenarios where users may revisit previously enjoyed content. We implemented and evaluated three distinct reward functions—focusing on maximizing group satisfaction, balancing satisfaction with diversity, and maximizing diversity—to investigate their impact on recommendation diversity and group satisfaction. Experimental results, using the 20M MovieLens dataset, demonstrate that while diversity-focused rewards enhance recommendation variety, it requires careful balance with user satisfaction to avoid diminishing user engagement.
KW - Diversity
KW - Group Recommendations
U2 - 10.1007/978-3-031-70421-5_24
DO - 10.1007/978-3-031-70421-5_24
M3 - Conference contribution
AN - SCOPUS:85210164112
SN - 9783031704208
T3 - Communications in Computer and Information Science
SP - 286
EP - 298
BT - New Trends in Database and Information Systems
A2 - Tekli, Joe
A2 - Gamper, Johann
A2 - Chbeir, Richard
A2 - Sassi, Salma
A2 - Manolopoulos, Yannis
A2 - Ivanovic, Mirjana
A2 - Vargas-Solar, Genoveva
A2 - Zumpano, Ester
PB - Springer
T2 - European Conference on Advances in Databases and Information Systems
Y2 - 28 August 2024 through 31 August 2024
ER -