TY - JOUR
T1 - Worship prediction
T2 - identify followers in celebrity-dived networks
AU - Teng, Shan Yun
AU - Ting, Lo Pang Yun
AU - Yeh, Mi Yen
AU - Chuang, Kun Ta
N1 - Funding Information:
Acknowledgements This study was supported in part by the Ministry of Science and Technology (MOST) of Taiwan, R.O.C., under Contracts 104-2628-E-001-005-MY3, 105-2628-E-001-002-MY2, 106-3114-E-002-008, and 105-2221-E-006-140-MY2. All opinions, findings, conclusions, and recommendations in this paper are those of the authors and do not necessarily reflect the views of the funding agencies.
Publisher Copyright:
© 2018, Springer Science+Business Media, LLC, part of Springer Nature.
PY - 2019/1/15
Y1 - 2019/1/15
N2 - We in this paper explore a new link prediction paradigm, called ‘worship’ prediction, to discover worship links between users and celebrities on social networks. The prediction of ‘worship’ links enables valuable social services, such as viral marketing, popularity estimation, and celebrity recommendation. However, as the concern of business security and personal privacy, only public-accessible statistical social properties, instead of the detailed information of users, can be utilized to predict the ‘worship’ labels. In addition, we observe that friendship properties are not effective to predict the desired links, meaning that most of previous work which rely on the friendship properties cannot be successfully applied in the prediction of worship link. To address these issues, a novel learning framework is devised, including a factor graph with new discovered statistical properties and a Gaussian estimation based learning algorithm with active learning. Our experimental studies on real data, including Instagram, Twitter and DBLP, show that the proposed learning framework can overcome the problem of missing labels and efficiently discover worship links.
AB - We in this paper explore a new link prediction paradigm, called ‘worship’ prediction, to discover worship links between users and celebrities on social networks. The prediction of ‘worship’ links enables valuable social services, such as viral marketing, popularity estimation, and celebrity recommendation. However, as the concern of business security and personal privacy, only public-accessible statistical social properties, instead of the detailed information of users, can be utilized to predict the ‘worship’ labels. In addition, we observe that friendship properties are not effective to predict the desired links, meaning that most of previous work which rely on the friendship properties cannot be successfully applied in the prediction of worship link. To address these issues, a novel learning framework is devised, including a factor graph with new discovered statistical properties and a Gaussian estimation based learning algorithm with active learning. Our experimental studies on real data, including Instagram, Twitter and DBLP, show that the proposed learning framework can overcome the problem of missing labels and efficiently discover worship links.
UR - http://www.scopus.com/inward/record.url?scp=85046042271&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85046042271&partnerID=8YFLogxK
U2 - 10.1007/s11280-018-0569-y
DO - 10.1007/s11280-018-0569-y
M3 - Article
AN - SCOPUS:85046042271
SN - 1386-145X
VL - 22
SP - 347
EP - 373
JO - World Wide Web
JF - World Wide Web
IS - 1
ER -