Filters








3 Hits in 2.1 sec

HoloScope: Topology-and-Spike Aware Fraud Detection [article]

Shenghua Liu, Bryan Hooi, Christos Faloutsos
2017 arXiv   pre-print
Hence, we propose HoloScope, which uses information from graph topology and temporal spikes to more accurately detect groups of fraudulent users.  ...  Extensive experiments showed that HoloScope achieved significant accuracy improvements on synthetic and real data, compared with state-of-the-art fraud detection methods.  ...  Our HoloScope metric detects fraud from three perspectives: topology connection, timestamps, and rating score.  ... 
arXiv:1705.02505v2 fatcat:rgumngio75g4tez4xzag4mpyqi

Blackmarket-driven Collusion on Online Media: A Survey [article]

Hridoy Sankar Dutta, Tanmoy Chakraborty
2020 arXiv   pre-print
We believe that collusive entity detection is a newly emerging topic in anomaly detection and cyber-security research in general and the current survey will provide readers with an easy-to-access and comprehensive  ...  list of methods, tools and resources proposed so far for detecting and analyzing collusive entities on online media.  ...  [83] formulated a novel suspiciousness metric from graph topology and spikes to detect fraudulent users in multiple rating platforms (BookAdvocate, Amazon). e authors proposed HoloScope, an unsupervised  ... 
arXiv:2008.13102v1 fatcat:vi6yiw5u7rbtvezi6fg32vcmwi

Blackmarket-Driven Collusion on Online Media: A Survey

Hridoy Sankar Dutta, Tanmoy Chakraborty
2021 ACM/IMS Transactions on Data Science  
Other than connectivity, these platforms also serve multiple purposes, such as education, promotion, updates, and awareness.  ...  We believe that collusive entity detection is a newly emerging topic in anomaly detection and cyber-security research in general, and the current survey will provide readers with an easy-to-access and  ...  [92] proposed HoloScope to find a group of suspicious users in rating platforms based on the contrasting behavior of fraudsters and honest users in terms of topology, temporal spikes, and rating deviation  ... 
doi:10.1145/3517931 fatcat:7fvgujegh5hohdiemsok6kzviq