FRAUDAR: Bounding Graph Fraud in the Face of Camouflage

@article{Hooi2016FRAUDARBG,
  title={FRAUDAR: Bounding Graph Fraud in the Face of Camouflage},
  author={Bryan Hooi and Hyun Ah Song and Alex Beutel and Neil Shah and Kijung Shin and Christos Faloutsos},
  journal={Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining},
  year={2016}
}
  • Bryan Hooi, H. Song, C. Faloutsos
  • Published 13 August 2016
  • Computer Science
  • Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining
Given a bipartite graph of users and the products that they review, or followers and followees, how can we detect fake reviews or follows? Existing fraud detection methods (spectral, etc.) try to identify dense subgraphs of nodes that are sparsely connected to the remaining graph. Fraudsters can evade these methods using camouflage, by adding reviews or follows with honest targets so that they look "normal". Even worse, some fraudsters use hijacked accounts from honest users, and then the… 

Figures and Tables from this paper

Camouflage is NOT easy: Uncovering adversarial fraudsters in large online app review platform
TLDR
This paper tackles the challenge of adversarial fraudsters in online app review platform and proposes a system called DDF (Detect, Defense, and Forecast) to uncover camouflage accounts and validate the efficiency and scalability of DDF and show that it can be well transferred to other anti-spam tasks.
GANG: Detecting Fraudulent Users in Online Social Networks via Guilt-by-Association on Directed Graphs
TLDR
GANG, a guilt-by-association method on directed graphs, to detect fraudulent users in OSNs is proposed and substantially outperforms existing methods, and the optimized version of GANG is significantly more efficient than the basic version.
A Contrast Metric for Fraud Detection in Rich Graphs
TLDR
HoloScope introduces a novel metric “contrast suspiciousness” integrating information from graph topology and spikes to more accurately detect fraudulent users and objects, and provides theoretical bounds for how much this method increases the time cost needed for fraudsters to conduct adversarial attacks.
Graph Fraud Detection Based on Accessibility Score Distributions
TLDR
SkewA, a novel fraud detection method that measures the skewness in accessibility score distributions and uses it as an honesty metric, is proposed and effective: showing up to 95.6% accuracy in real-world data where all competitors fail to detect any fraud.
Fraud De-Anonymization for Fun and Profit
TLDR
This work develops a graph based deep learning approach to predict ownership of account pairs by the same fraudster and uses it to build discriminative fraud de-anonymization (DDA) and pseudonymous fraudster discovery algorithms (PFD).
Fraud Detection using Graph Topology and Temporal Spikes
TLDR
A novel “contrast suspiciousness” is introduced for honoring contrast behaviors between fraudsters and normal users and HoloScope has a concise framework and sub-quadratic time complexity, which make it reproducible and scalable.
FdGars: Fraudster Detection via Graph Convolutional Networks in Online App Review System
TLDR
The first work to use graph convolutional networks for fraudster detection in the large-scale online app review system and it is worth to mention that FdGars can uncover malicious accounts even the data lack of labels in anti-spam tasks.
Search Rank Fraud De-Anonymization in Online Systems
TLDR
Dolos, a fraud de-anonymization system that leverages traits and behaviors extracted from studies, to attribute detected fraud to crowdsourcing site fraudsters, thus to real identities and bank accounts is proposed.
Collusion-aware detection of review spammers in location based social networks
TLDR
This paper proposes a novel detection mechanism to discern collusive review spammers, including individuals and groups, and shows that the method can detect collusive spammers and groups more accurately and comprehensively over the current researches.
One-Class Adversarial Nets for Fraud Detection
TLDR
Experimental results show that the OCAN outperforms the state-of-the-art one-class classification models and achieves comparable performance with the latest multi-source LSTM model that requires both benign and malicious users in the training phase.
...
1
2
3
4
5
...

References

SHOWING 1-10 OF 37 REFERENCES
Opinion Fraud Detection in Online Reviews by Network Effects
TLDR
A fast and effective framework for spotting fraudsters and fake reviews in online review datasets, FRAUDEAGLE, which exploits the network effect among reviewers and products and is scalable to large datasets as its run time grows linearly with network size.
Spotting Suspicious Link Behavior with fBox: An Adversarial Perspective
TLDR
This work takes an adversarial approach to find and prove claims about the weaknesses of modern, state-of-the-art spectral methods and proposes fBox, an algorithm designed to catch small-scale, stealth attacks that slip below the radar.
CopyCatch: stopping group attacks by spotting lockstep behavior in social networks
TLDR
The method, which is referred to as CopyCatch, detects lockstep Page Like patterns on Facebook by analyzing only the social graph between users and Pages and the times at which the edges in the graph were created.
Aiding the Detection of Fake Accounts in Large Scale Social Online Services
TLDR
A new tool in the hands of OSN operators, which relies on social graph properties to rank users according to their perceived likelihood of being fake (SybilRank), which is computationally efficient and can scale to graphs with hundreds of millions of nodes, as demonstrated by the Hadoop prototype.
Netprobe: a fast and scalable system for fraud detection in online auction networks
TLDR
This paper describes the design and implementation of NetProbe, a system that is a Markov Random Field tuned to detect the suspicious patterns that fraudsters create, and employs a Belief Propagation mechanism to detect likely fraudsters.
Sybil-Resilient Online Content Voting
TLDR
SumUp is presented, a Sybilresilient vote aggregation system that leverages the trust network among users to defend against Sybil attacks and uses the technique of adaptive vote flow aggregation to limit the number of bogus votes cast by adversaries to no more than theNumber of attack edges in the Trust network.
Camouflaged fraud detection in domains with complex relationships
TLDR
A data mining system to detect frauds that are camouflaged to look like normal activities in domains with high number of known relationships to help analysts who are overwhelmed with information about companies or on-line system access logs or insurance claims to focus their attentions on features that cause damage in the future.
SybilGuard: Defending Against Sybil Attacks via Social Networks
TLDR
This paper presents SybilGuard, a novel protocol for limiting the corruptive influences of sybil attacks, based on the ldquosocial networkrdquo among user identities, where an edge between two identities indicates a human-established trust relationship.
CatchSync: catching synchronized behavior in large directed graphs
TLDR
This work proposes a fast and effective method, CatchSync, which exploits two of the tell-tale signs left in graphs by fraudsters, and introduces novel measures to quantify both concepts ("synchronicity" and "normality") and proposes a parameter-free algorithm that works on the resulting synchronicities-normality plots.
A General Suspiciousness Metric for Dense Blocks in Multimodal Data
TLDR
This work gives a list of axioms that any metric of suspicousness should satisfy, proposes CROSSSPOT, an algorithm to spot dense regions, and sort them in importance ("suspiciousness") order, and applies it to real data, where it improves the F1 score over previous techniques by 68% and finds retweet-boosting in a real social dataset spanning 0.3 billion posts.
...
1
2
3
4
...