Distance-based customer detection in fake follower markets

作者：

Highlights：

• Link probability of normal in online social network is relevant to geographical distance.

• Abnormal users show different pattern in terms of link probability.

• Customers of fake followers have different follower ratio from legitimate users.

• Combining geographical distance and relationship is efficient to detect customers.

摘要

•Link probability of normal in online social network is relevant to geographical distance.•Abnormal users show different pattern in terms of link probability.•Customers of fake followers have different follower ratio from legitimate users.•Combining geographical distance and relationship is efficient to detect customers.

论文关键词：Online social networks,Twitter,Fake follower,Follower spam,Geographical distance,Small-world phenomenon

论文评审过程：Received 15 May 2018, Revised 15 October 2018, Accepted 3 December 2018, Available online 7 December 2018, Version of Record 20 December 2018.

论文官网地址：https://doi.org/10.1016/j.is.2018.12.001