TY - JOUR
T1 - Improving stability of recommender systems
T2 - A meta-algorithmic approach
AU - Adomavicius, Gediminas
AU - Zhang, Jingjing
PY - 2015/6/1
Y1 - 2015/6/1
N2 - This paper focuses on the measure of recommendation stability, which reflects the consistency of recommender system predictions. Stability is a desired property of recommendation algorithms and has important implications on users' trust and acceptance of recommendations. Prior research has reported that some popular recommendation algorithms can suffer from a high degree of instability. In this study, we explore two scalable, general-purpose meta-algorithmic approaches - based on bagging and iterative smoothing - that can be used in conjunction with different traditional recommendation algorithms to improve their stability. Our experimental results on real-world rating data demonstrate that both approaches can achieve substantially higher stability as compared to the original recommendation algorithms. Furthermore, perhaps as importantly, the proposed approaches not only do not sacrifice the predictive accuracy in order to improve recommendation stability, but are actually able to provide additional accuracy improvements.
AB - This paper focuses on the measure of recommendation stability, which reflects the consistency of recommender system predictions. Stability is a desired property of recommendation algorithms and has important implications on users' trust and acceptance of recommendations. Prior research has reported that some popular recommendation algorithms can suffer from a high degree of instability. In this study, we explore two scalable, general-purpose meta-algorithmic approaches - based on bagging and iterative smoothing - that can be used in conjunction with different traditional recommendation algorithms to improve their stability. Our experimental results on real-world rating data demonstrate that both approaches can achieve substantially higher stability as compared to the original recommendation algorithms. Furthermore, perhaps as importantly, the proposed approaches not only do not sacrifice the predictive accuracy in order to improve recommendation stability, but are actually able to provide additional accuracy improvements.
KW - Recommender systems
KW - bagging
KW - collaborative filtering
KW - iterative smoothing
KW - recommendation stability
UR - http://www.scopus.com/inward/record.url?scp=84929459410&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84929459410&partnerID=8YFLogxK
U2 - 10.1109/TKDE.2014.2384502
DO - 10.1109/TKDE.2014.2384502
M3 - Article
AN - SCOPUS:84929459410
VL - 27
SP - 1573
EP - 1587
JO - IEEE Transactions on Knowledge and Data Engineering
JF - IEEE Transactions on Knowledge and Data Engineering
SN - 1041-4347
IS - 6
M1 - 6994303
ER -