TY - GEN
T1 - Modeling the Changing of the Individual Satisfaction in a Group Context
T2 - 17th Conference of the Italian Association for Artificial Intelligence, AI*IA 2018
AU - Barile, Francesco
AU - Masthoff, Judith
AU - Rossi, Silvia
PY - 2018/1/1
Y1 - 2018/1/1
N2 - General approaches for Group Recommendation Systems start from the individual recommendations and merge them in a way to determine the best choice for the whole group. The results presented in the literature showed that traditional aggregation techniques do not seem to capture all the features of real-world scenarios. Furthermore, recent studies in Behavioral Economics evidence the necessity to define utility models that are not compatible with the self-interested utility-maximizing behavior of the traditional economic paradigm. In this work, starting from Other-Regarding Preference models that characterize the utility of an individual considering his/her own behavioral characteristics and the utility of another individual, we aim at obtaining a general model where such characteristics are described in terms of interpersonal relationships, such as, tie strength and conflict. We started by performing an analysis on opinion shifts based on a two sized groups user study, with the aim to empirically determine the extent of the considered parameters on a possible model. The results show that the opinion shifting on the evaluation of an activity to be performed in a group is related to the two considered factors.
AB - General approaches for Group Recommendation Systems start from the individual recommendations and merge them in a way to determine the best choice for the whole group. The results presented in the literature showed that traditional aggregation techniques do not seem to capture all the features of real-world scenarios. Furthermore, recent studies in Behavioral Economics evidence the necessity to define utility models that are not compatible with the self-interested utility-maximizing behavior of the traditional economic paradigm. In this work, starting from Other-Regarding Preference models that characterize the utility of an individual considering his/her own behavioral characteristics and the utility of another individual, we aim at obtaining a general model where such characteristics are described in terms of interpersonal relationships, such as, tie strength and conflict. We started by performing an analysis on opinion shifts based on a two sized groups user study, with the aim to empirically determine the extent of the considered parameters on a possible model. The results show that the opinion shifting on the evaluation of an activity to be performed in a group is related to the two considered factors.
KW - Group Recommendation Systems
KW - Other Regarding Preferences
KW - Social dynamics
UR - http://www.scopus.com/inward/record.url?scp=85057416973&partnerID=8YFLogxK
U2 - 10.1007/978-3-030-03840-3_36
DO - 10.1007/978-3-030-03840-3_36
M3 - Conference contribution
AN - SCOPUS:85057416973
SN - 9783030038397
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 489
EP - 501
BT - AI*IA 2018 – Advances in Artificial Intelligence - 17th International Conference of the Italian Association for Artificial Intelligence, Proceedings
A2 - Ghidini, Chiara
A2 - Traverso, Paolo
A2 - Magnini, Bernardo
A2 - Passerini, Andrea
PB - Springer
Y2 - 20 November 2018 through 23 November 2018
ER -