TY - GEN
T1 - Similar but different
T2 - 11th International Conference on Social Computing, Behavioral-Cultural Modeling, and Prediction conference and Behavior Representation in Modeling and Simulation, SBP-BRiMS 2018
AU - Beigi, Ghazaleh
AU - Liu, Huan
N1 - Funding Information: Acknowledgments. This material is based upon the work supported by, or in part by, Office of Naval Research (ONR) under grant number N00014-17-1-2605. Publisher Copyright: © 2018, Springer International Publishing AG, part of Springer Nature.
PY - 2018
Y1 - 2018
N2 - The pervasive use of social media provides massive data about individuals’ online social activities and their social relations. The building block of most existing recommendation systems is the similarity between users with social relations, i.e., friends. While friendship ensures some homophily, the similarity of a user with her friends can vary as the number of friends increases. Research from sociology suggests that friends are more similar than strangers, but friends can have different interests. Exogenous information such as comments and ratings may help discern different degrees of agreement (i.e., congruity) among similar users. In this paper, we investigate if users’ congruity can be incorporated into recommendation systems to improve it’s performance. Experimental results demonstrate the effectiveness of embedding congruity related information into recommendation systems.
AB - The pervasive use of social media provides massive data about individuals’ online social activities and their social relations. The building block of most existing recommendation systems is the similarity between users with social relations, i.e., friends. While friendship ensures some homophily, the similarity of a user with her friends can vary as the number of friends increases. Research from sociology suggests that friends are more similar than strangers, but friends can have different interests. Exogenous information such as comments and ratings may help discern different degrees of agreement (i.e., congruity) among similar users. In this paper, we investigate if users’ congruity can be incorporated into recommendation systems to improve it’s performance. Experimental results demonstrate the effectiveness of embedding congruity related information into recommendation systems.
UR - http://www.scopus.com/inward/record.url?scp=85049780429&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85049780429&partnerID=8YFLogxK
U2 - 10.1007/978-3-319-93372-6_15
DO - 10.1007/978-3-319-93372-6_15
M3 - Conference contribution
SN - 9783319933719
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 129
EP - 140
BT - Social, Cultural, and Behavioral Modeling - 11th International Conference, SBP-BRiMS 2018, Proceedings
A2 - Bisgin, Halil
A2 - Thomson, Robert
A2 - Hyder, Ayaz
A2 - Dancy, Christopher
PB - Springer Verlag
Y2 - 10 July 2018 through 13 July 2018
ER -