Collaborative User Network Embedding for Social Recommender Systems

Handle URI:
http://hdl.handle.net/10754/625053
Title:
Collaborative User Network Embedding for Social Recommender Systems
Authors:
Zhang, Chuxu; Yu, Lu; Wang, Yan; Shah, Chirag; Zhang, Xiangliang ( 0000-0002-3574-5665 )
Abstract:
To address the issue of data sparsity and cold-start in recommender system, social information (e.g., user-user trust links) has been introduced to complement rating data for improving the performances of traditional model-based recommendation techniques such as matrix factorization (MF) and Bayesian personalized ranking (BPR). Although effective, the utilization of the explicit user-user relationships extracted directly from such social information has three main limitations. First, it is difficult to obtain explicit and reliable social links. Only a small portion of users indicate explicitly their trusted friends in recommender systems. Second, the “cold-start” users are “cold” not only on rating but also on socializing. There is no significant amount of explicit social information that can be useful for “cold-start” users. Third, an active user can be socially connected with others who have different taste/preference. Direct usage of explicit social links may mislead recommendation. To address these issues, we propose to extract implicit and reliable social information from user feedbacks and identify top-k semantic friends for each user. We incorporate the top-k semantic friends information into MF and BPR frameworks to solve the problems of ratings prediction and items ranking, respectively. The experimental results on three real-world datasets show that our proposed approaches achieve better results than the state-of-the-art MF with explicit social links (with 3.0% improvement on RMSE), and social BPR (with 9.1% improvement on AUC).
KAUST Department:
Computer, Electrical and Mathematical Sciences and Engineering (CEMSE) Division; Computer Science Program
Citation:
Zhang C, Yu L, Wang Y, Shah C, Zhang X (2017) Collaborative User Network Embedding for Social Recommender Systems. Proceedings of the 2017 SIAM International Conference on Data Mining: 381–389. Available: http://dx.doi.org/10.1137/1.9781611974973.43.
Publisher:
Society for Industrial and Applied Mathematics
Journal:
Proceedings of the 2017 SIAM International Conference on Data Mining
Issue Date:
9-Jun-2017
DOI:
10.1137/1.9781611974973.43
Type:
Book Chapter
Sponsors:
Research reported in this publication was partially supported by the US Institute of Museum and Library Services (IMLS) National Leadership Grant #LG-81-16-0025, and the King Abdullah University of Science and Technology (KAUST).
Additional Links:
http://epubs.siam.org/doi/10.1137/1.9781611974973.43
Appears in Collections:
Computer Science Program; Book Chapters; Computer, Electrical and Mathematical Sciences and Engineering (CEMSE) Division

Full metadata record

DC FieldValue Language
dc.contributor.authorZhang, Chuxuen
dc.contributor.authorYu, Luen
dc.contributor.authorWang, Yanen
dc.contributor.authorShah, Chiragen
dc.contributor.authorZhang, Xiangliangen
dc.date.accessioned2017-06-19T09:21:45Z-
dc.date.available2017-06-19T09:21:45Z-
dc.date.issued2017-06-09en
dc.identifier.citationZhang C, Yu L, Wang Y, Shah C, Zhang X (2017) Collaborative User Network Embedding for Social Recommender Systems. Proceedings of the 2017 SIAM International Conference on Data Mining: 381–389. Available: http://dx.doi.org/10.1137/1.9781611974973.43.en
dc.identifier.doi10.1137/1.9781611974973.43en
dc.identifier.urihttp://hdl.handle.net/10754/625053-
dc.description.abstractTo address the issue of data sparsity and cold-start in recommender system, social information (e.g., user-user trust links) has been introduced to complement rating data for improving the performances of traditional model-based recommendation techniques such as matrix factorization (MF) and Bayesian personalized ranking (BPR). Although effective, the utilization of the explicit user-user relationships extracted directly from such social information has three main limitations. First, it is difficult to obtain explicit and reliable social links. Only a small portion of users indicate explicitly their trusted friends in recommender systems. Second, the “cold-start” users are “cold” not only on rating but also on socializing. There is no significant amount of explicit social information that can be useful for “cold-start” users. Third, an active user can be socially connected with others who have different taste/preference. Direct usage of explicit social links may mislead recommendation. To address these issues, we propose to extract implicit and reliable social information from user feedbacks and identify top-k semantic friends for each user. We incorporate the top-k semantic friends information into MF and BPR frameworks to solve the problems of ratings prediction and items ranking, respectively. The experimental results on three real-world datasets show that our proposed approaches achieve better results than the state-of-the-art MF with explicit social links (with 3.0% improvement on RMSE), and social BPR (with 9.1% improvement on AUC).en
dc.description.sponsorshipResearch reported in this publication was partially supported by the US Institute of Museum and Library Services (IMLS) National Leadership Grant #LG-81-16-0025, and the King Abdullah University of Science and Technology (KAUST).en
dc.publisherSociety for Industrial and Applied Mathematicsen
dc.relation.urlhttp://epubs.siam.org/doi/10.1137/1.9781611974973.43en
dc.rightsArchived with thanks to Proceedings of the 2017 SIAM International Conference on Data Miningen
dc.subjectSocial Recommender Systemsen
dc.subjectNetwork Embeddingen
dc.subjectTop-k Semantic Friendsen
dc.subjectMatrix Factorizationen
dc.subjectBayesian Personalized Rankingen
dc.titleCollaborative User Network Embedding for Social Recommender Systemsen
dc.typeBook Chapteren
dc.contributor.departmentComputer, Electrical and Mathematical Sciences and Engineering (CEMSE) Divisionen
dc.contributor.departmentComputer Science Programen
dc.identifier.journalProceedings of the 2017 SIAM International Conference on Data Miningen
dc.eprint.versionPublisher's Version/PDFen
dc.contributor.institutionRutgers University, USA.en
kaust.authorYu, Luen
kaust.authorZhang, Xiangliangen
All Items in KAUST are protected by copyright, with all rights reserved, unless otherwise indicated.