TY - GEN
T1 - 'Can you really trust that seed?'
T2 - 2014 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining, ASONAM 2014
AU - Huang, Shengyu
AU - Li, Xinsheng
AU - Candan, Kasim
AU - Sapino, Maria Luisa
N1 - Publisher Copyright:
© 2014 IEEE.
PY - 2014/10/10
Y1 - 2014/10/10
N2 - Network based recommendation systems leverage the topology of the underlying graph and the current user context to rank objects in the database. Random-walk based techniques, such as PageRank, encode the structure of the graph in the form of a transition matrix of a stochastic process from which the significances of the nodes in the graph are inferred. Personalized PageRank (PPR) techniques complement this with a seed node set which serves as the personalization context. In this paper, we note (and experimentally show) that PPR algorithms that do not differentiate among the seed nodes may not properly rank nodes in situations where the seed set is incomplete and/or noisy. To tackle this problem, we propose alternative robust personalized PageRank (RPR) strategies, which are insensitive to noise in the set of seed nodes and in which the rankings are not overly biased towards the seed nodes. In particular, we show that novel teleportation discounting and seed-set maximal PPR techniques help eliminate harmful bias of individual seed nodes and provide effective seed differentiation to lead to more accurate rankings.
AB - Network based recommendation systems leverage the topology of the underlying graph and the current user context to rank objects in the database. Random-walk based techniques, such as PageRank, encode the structure of the graph in the form of a transition matrix of a stochastic process from which the significances of the nodes in the graph are inferred. Personalized PageRank (PPR) techniques complement this with a seed node set which serves as the personalization context. In this paper, we note (and experimentally show) that PPR algorithms that do not differentiate among the seed nodes may not properly rank nodes in situations where the seed set is incomplete and/or noisy. To tackle this problem, we propose alternative robust personalized PageRank (RPR) strategies, which are insensitive to noise in the set of seed nodes and in which the rankings are not overly biased towards the seed nodes. In particular, we show that novel teleportation discounting and seed-set maximal PPR techniques help eliminate harmful bias of individual seed nodes and provide effective seed differentiation to lead to more accurate rankings.
UR - http://www.scopus.com/inward/record.url?scp=84911095295&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84911095295&partnerID=8YFLogxK
U2 - 10.1109/ASONAM.2014.6921586
DO - 10.1109/ASONAM.2014.6921586
M3 - Conference contribution
AN - SCOPUS:84911095295
T3 - ASONAM 2014 - Proceedings of the 2014 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining
SP - 216
EP - 223
BT - ASONAM 2014 - Proceedings of the 2014 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining
A2 - Wu, Xindong
A2 - Wu, Xindong
A2 - Ester, Martin
A2 - Xu, Guandong
PB - Institute of Electrical and Electronics Engineers Inc.
Y2 - 17 August 2014 through 20 August 2014
ER -