A PERSONALITY-BASED AGGREGATION TECHNIQUE FOR GROUP RECOMMENDATION
Abstract
The main goal of a group recommender system is to provide appropriate referrals to a group of users sharing common interests rather than individuals. Such group referrals are commonly produced by utilizing aggregation techniques that analyze the propensities of the whole group by combining the preferences of the users in the group. Although there exist various aggregation techniques in the literature, they usually rely on the assumption that each member of the group has equal importance on the final decision of the group. However, the decision-making process of a group is a complicated process that is strongly correlated with not only group members' experience about the domain of interest but also their behavioral aspects; therefore, the influence of the individuals might be dependent on user personalities. In this study, we propose a personality-aware aggregation technique termed as the Personality weighted Average (PwAvg), which determines the influence degree of each member in the group using five fundamental personality traits, openness, agreeableness, emotional stability, conscientiousness, and extraversion; and then utilizes them to weight the preferences during the aggregation process. Experiments performed on two real-world benchmark datasets demonstrate that the PwAvg technique significantly outperforms three baseline aggregation techniques, especially for large user groups. Empirical outcomes also show that utilizing the PwAvg with emotional stability trait achieves more qualified group recommendations compared to others.
Keywords
Group recommender system, aggregation techniques, personality traits
References
- [1] Adomavicius G, Tuzhilin A. Toward the next generation of recommender systems: A survey of the state-of-the-art and possible extensions. IEEE transactions on knowledge and data engineering 2005; 17(6): 734-749.
- [2] Ricci F, Rokach L, Shapira B. Introduction to recommender systems handbook. Springer US, 2011.
- [3] Goren-Bar D, Glinansky O. FIT-recommending TV programs to family members. Computers & Graphics 2004; 28.2: 149-156.
- [4] McCarthy JF. Pocket restaurantfinder: A situated recommender system for groups. In: Workshop on Mobile Ad-Hoc Communication at the 2002 ACM Conference on Human Factors in Computer Systems, 2002.
- [5] Jameson A. More than the sum of its members: challenges for group recommender systems. In: Proceedings of the working conference on Advanced visual interfaces, 2004, 48-54.
- [6] McCarthy K, Salamó M, Coyle L, McGinty L, Smyth B, Nixon P. Cats: A synchronous approach to collaborative group recommendation. In: Florida Artificial Intelligence Research Society Conference (FLAIRS), 2006, 86-91.
- [7] Ardissono L, Goy A, Petrone G, Segnan M, Torasso P. Intrigue: personalized recommendation of tourist attractions for desktop and hand held devices. Applied artificial intelligence 2003; 17(8-9): 687-714.
- [8] McCarthy JF, Anagnost TD. MusicFX: an arbiter of group preferences for computer supported collaborative workouts. In: Proceedings of the 1998 ACM conference on computer supported cooperative work, 1998, 363-372.
- [9] Jameson A. More than the sum of its members: challenges for group recommender systems. In: Proceedings of the working conference on Advanced visual interfaces, 2004, 48-54.
- [10] Chao DL, Balthrop J, Forrest S. Adaptive radio: achieving consensus using negative preferences. In: Proceedings of the 2005 international ACM SIGGROUP conference on Supporting group work, 2005, 120-123.