Combating Crowdsourced Review Manipulators: A Neighborhood-Based Approach
Open accessAbstract
We propose a system called TwoFace to uncover crowdsourced review manipulators who target online review systems. A unique feature of TwoFace is its three-phase framework:(i) in the first phase, we intelligently sample actual evidence of manipulation(e.g., review manipulators) by exploiting low moderation crowdsourcing platforms that reveal evidence of strategic manipulation;(ii) we then propagate the suspiciousness of these seed users to identify similar users through a random walk over a "suspiciousness»» graph; and(iii) finally, we uncover(hidden) distant users who serve structurally similar roles by mapping users into a low-dimensional embedding space that captures community structure. Altogether, the TwoFace system recovers 83% to 93% of all manipulators in a sample from Amazon of 38,590 reviewers, even when the system is seeded with only a few samples from malicious crowdsourcing sites.
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
The paper found rapid review bursts in both qualitative and quantitative investigations, consistent with assumptions in prior fraud-detection work, and interview participants revealed adaptive rate control as an avoidance technique.
“We found ASO workers who post fake reviews in rapid bursts in both our qualitative and quantitative investigations.This is consistent with assumptions made in previous fraud detection work, e.g., [26, 28, 32, 36-38, 40, 42, 43, 51, 52, 59-61, 63, 65, 92, 95, 96, 100].”
From The Art and Craft of Fraudulent App Promotion in Google Play (2019) · cites this paper as [26]
Because multiple fraud workers usually target one subject, binary classification of fraud is insufficient to understand and model fraud.
“The insight that multiple fraud workers usually target a single subject suggests that a binary classification of fraud, e.g., fake vs. honest reviews, fraudulent vs. honest accounts [17,26,27,31,32,44,47,80], is insufficient to understand and model fraud.”
From Fraud De-Anonymization for Fun and Profit (2018) · cites this paper as [26]
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
The citation lists haven’t been fetched yet
References 34 · Cited by 59
Research works printed but not linked: 3
- Bryan Hooi, Neil Shah, Alex Beutel, Stephan Günnemann, Leman Akoglu, Mohit Kumar, Disha Makhija, and Christos Faloutsos. 2016. Birdnest: Bayesian infer- ence for ratings-fraud detection. In Proceedings of the 2016 SIAM International Conference on Data Mining. SIAM, 495-503.
- Myle Ott, Yejin Choi, Claire Cardie, and Jeffrey T Hancock. 2011. Finding decep- tive opinion spam by any stretch of the imagination. In ACL.
- Kijung Shin, Bryan Hooi, Jisu Kim, and Christos Faloutsos. 2017. D-cube: Dense- block detection in terabyte-scale tensors. In Proceedings of the Tenth ACM Inter- national Conference on Web Search and Data Mining. ACM, 681-689.
- Authors
- Kaghazgaran, P., Caverlee, J., & Squicciarini, A.
- Affiliations
- Texas A&M University (US)
- Pennsylvania State University (US)
- Venue
- Proceedings of the Eleventh ACM International Conference on Web Search and Data Mining
- Publisher
- ACM
- Published
- 2018-02-02
- Pages
- 306-314
- OpenAlex
- W2782836818
- Cited by
- 59
- References
- 34
- Citations by year
- 2026: 3 · 2025: 4 · 2024: 7 · 2023: 5 · 2022: 7 · 2021: 10 · 2020: 12 · 2019: 9 · 2018: 2
- FWCI
- 6.47
- Type
- conference paper
- Language
- en
- Open access
- bronze
- doi.org
- Locations
- ACM International Conference on Web Search and Data Mining (WSDM) (publishedVersion)
- Funders
- AFOSR (FA9550-15-1-0149)
- Topics
- Physical Sciences › Computer Science › Information Systems › Spam and Phishing Detection
- Social Sciences › Sociology and Political Science › Misinformation and Its Impacts
- Physical Sciences › Physics and Astronomy › Statistical and Nonlinear Physics › Complex Network Analysis Techniques
- Keywords
- online reviews, spam review detection, user profiling, random walks, community structure, user embeddings, Amazon reviews
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
