TY - GEN
T1 - SocialDistance
T2 - 2019 International Symposium on Quality of Service, IWQoS 2019
AU - Li, Ang
AU - Li, Tao
AU - Zhang, Yan
AU - Zhang, Lili
AU - Zhang, Yanchao
N1 - Funding Information:
We would like to thank the anonymous reviewers for their insightful comments that help improve the quality of this paper. This work was supported in part by Army Research Office under grant W911NF-15-1-0328 and US National Science Foundation under grants CNS-1514381, CNS-1619251, and CNS-1824355.
Publisher Copyright:
© 2019 Association for Computing Machinery.
PY - 2019/6/24
Y1 - 2019/6/24
N2 - Verified users on online social media (OSM) largely determine the quality of OSM services and applications, but most OSM users are unverified due to the significant effort involved in becoming a verified user. This paper presents SocialDistance, a novel technique to identify unverified users that can be considered as trustworthy as verified users. SocialDistance is motivated by the observation that online interactions initiated from verified users towards unverified users can translate into some sort of trustworthiness. It treats all verified users equally and assigns a trust score between 0 and 1 to each unverified user. The higher the trust score, the closer an unverified user to verified users. We propose various metrics to model the interactions from verified to unverified users and then derive corresponding trust scores. SocialDistance is thoroughly evaluated with large Twitter datasets containing 276,143 verified users and 19,047,202 unverified users. Our results demonstrate that SocialDistance can produce a non-trivial number of unverified users that can be regarded as verified users for OSM applications. We also show the high efficacy of SocialDistance in sybil detection, a fundamental operation performed by virtually every OSM operator.
AB - Verified users on online social media (OSM) largely determine the quality of OSM services and applications, but most OSM users are unverified due to the significant effort involved in becoming a verified user. This paper presents SocialDistance, a novel technique to identify unverified users that can be considered as trustworthy as verified users. SocialDistance is motivated by the observation that online interactions initiated from verified users towards unverified users can translate into some sort of trustworthiness. It treats all verified users equally and assigns a trust score between 0 and 1 to each unverified user. The higher the trust score, the closer an unverified user to verified users. We propose various metrics to model the interactions from verified to unverified users and then derive corresponding trust scores. SocialDistance is thoroughly evaluated with large Twitter datasets containing 276,143 verified users and 19,047,202 unverified users. Our results demonstrate that SocialDistance can produce a non-trivial number of unverified users that can be regarded as verified users for OSM applications. We also show the high efficacy of SocialDistance in sybil detection, a fundamental operation performed by virtually every OSM operator.
KW - Interaction graph
KW - Quality of online social media
KW - Sybil detection
KW - Unverified user
KW - Verified user
UR - http://www.scopus.com/inward/record.url?scp=85069190413&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85069190413&partnerID=8YFLogxK
U2 - 10.1145/3326285.3329075
DO - 10.1145/3326285.3329075
M3 - Conference contribution
AN - SCOPUS:85069190413
T3 - Proceedings of the International Symposium on Quality of Service, IWQoS 2019
BT - Proceedings of the International Symposium on Quality of Service, IWQoS 2019
PB - Association for Computing Machinery, Inc
Y2 - 24 June 2019 through 25 June 2019
ER -