STARS: Defending against Sockpuppet-Based Targeted Attacks on Reviewing Systems

Rui Liu; Runze Liu; Andrea Pugliese; V. S. Subrahmanian

doi:10.1145/3397463

STARS: Defending against Sockpuppet-Based Targeted Attacks on Reviewing Systems

Rui Liu, Runze Liu, Andrea Pugliese, V. S. Subrahmanian

Computer Science

Research output: Contribution to journal › Article › peer-review

11 Scopus citations

Abstract

Customers of virtually all online marketplaces rely upon reviews in order to select the product or service they wish to buy. These marketplaces in turn deploy review fraud detection systems so that the integrity of reviews is preserved. A well-known problem with review fraud detection systems is their underlying assumption that the majority of reviews are honest-This assumption leads to a vulnerability where an attacker can try to generate many fake reviews of a product. In this article, we consider the case where a company wishes to fraudulently promote its product through fake reviews and propose the Sockpuppet-based Targeted Attack on Reviewing Systems (STARS for short). STARS enables an attacker to enter fake reviews for a product from multiple, apparently independent, sockpuppet accounts. We show that the STARS attack enables companies to successfully promote their product against seven recent, well-known review fraud detectors on four datasets (Amazon, Epinions, and the BitcoinAlpha and OTC exchanges) by significant margins. To protect against the STARS attack, we propose a new fraud detection algorithm called RTV. RTV introduces a new class of users (called trusted users) and also considers reviews left by verified users which were not considered in existing review fraud detectors. We show that RTV significantly mitigates the impact of the STARS attack across the four datasets listed above.

Original language	English (US)
Article number	56
Journal	ACM Transactions on Intelligent Systems and Technology
Volume	11
Issue number	5
DOIs	https://doi.org/10.1145/3397463
State	Published - Sep 2020

Funding

Parts of this work were funded by ONR grants N00014-18-1-2670, N00014-16-1-2896, and N00014-20-1-2407. Parts of this work were funded by ONR grants N00014-18-1-2670, N00014-16-1-2896, and N00014-20-1-2407. Authors’ addresses: R. Liu and V. S. Subrahmanian, 6211 Sudikoff Lab, Dartmouth College, Hanover, NH 03755, USA; emails: {Rui.Liu.GR, vs}@dartmouth.edu; R. Liu, 1079 Commonwealth Avenue, APT 508, Boston, MA, 02215, USA; email: [email protected]; A. Pugliese, Cubo 44Z, University of Calabria, Rende, Italy; email: [email protected]. Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. Copyrights for components of this work owned by others than ACM must be honored. Abstracting with credit is permitted. To copy otherwise, or republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee. Request permissions from [email protected]. © 2020 Association for Computing Machinery. 2157-6904/2020/07-ART56 $15.00 https://doi.org/10.1145/3397463

Keywords

Data mining and knowledge discovery
online commerce and recommendation systems
social and information networks

ASJC Scopus subject areas

Theoretical Computer Science
Artificial Intelligence

Access to Document

10.1145/3397463

Cite this

@article{e42022a438c64f0d8c07071d2e6b3b54,

title = "STARS: Defending against Sockpuppet-Based Targeted Attacks on Reviewing Systems",

abstract = "Customers of virtually all online marketplaces rely upon reviews in order to select the product or service they wish to buy. These marketplaces in turn deploy review fraud detection systems so that the integrity of reviews is preserved. A well-known problem with review fraud detection systems is their underlying assumption that the majority of reviews are honest-This assumption leads to a vulnerability where an attacker can try to generate many fake reviews of a product. In this article, we consider the case where a company wishes to fraudulently promote its product through fake reviews and propose the Sockpuppet-based Targeted Attack on Reviewing Systems (STARS for short). STARS enables an attacker to enter fake reviews for a product from multiple, apparently independent, sockpuppet accounts. We show that the STARS attack enables companies to successfully promote their product against seven recent, well-known review fraud detectors on four datasets (Amazon, Epinions, and the BitcoinAlpha and OTC exchanges) by significant margins. To protect against the STARS attack, we propose a new fraud detection algorithm called RTV. RTV introduces a new class of users (called trusted users) and also considers reviews left by verified users which were not considered in existing review fraud detectors. We show that RTV significantly mitigates the impact of the STARS attack across the four datasets listed above.",

keywords = "Data mining and knowledge discovery, online commerce and recommendation systems, social and information networks",

author = "Rui Liu and Runze Liu and Andrea Pugliese and Subrahmanian, {V. S.}",

note = "Funding Information: Parts of this work were funded by ONR grants N00014-18-1-2670, N00014-16-1-2896, and N00014-20-1-2407. Funding Information: Parts of this work were funded by ONR grants N00014-18-1-2670, N00014-16-1-2896, and N00014-20-1-2407. Authors{\textquoteright} addresses: R. Liu and V. S. Subrahmanian, 6211 Sudikoff Lab, Dartmouth College, Hanover, NH 03755, USA; emails: {Rui.Liu.GR, vs}@dartmouth.edu; R. Liu, 1079 Commonwealth Avenue, APT 508, Boston, MA, 02215, USA; email: [email protected]; A. Pugliese, Cubo 44Z, University of Calabria, Rende, Italy; email: [email protected]. Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. Copyrights for components of this work owned by others than ACM must be honored. Abstracting with credit is permitted. To copy otherwise, or republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee. Request permissions from [email protected]. {\textcopyright} 2020 Association for Computing Machinery. 2157-6904/2020/07-ART56 $15.00 https://doi.org/10.1145/3397463 Publisher Copyright: {\textcopyright} 2020 ACM.",

year = "2020",

month = sep,

doi = "10.1145/3397463",

language = "English (US)",

volume = "11",

journal = "ACM Transactions on Intelligent Systems and Technology",

issn = "2157-6904",

publisher = "Association for Computing Machinery (ACM)",

number = "5",

}

TY - JOUR

T1 - STARS

T2 - Defending against Sockpuppet-Based Targeted Attacks on Reviewing Systems

AU - Liu, Rui

AU - Liu, Runze

AU - Pugliese, Andrea

AU - Subrahmanian, V. S.

N1 - Funding Information: Parts of this work were funded by ONR grants N00014-18-1-2670, N00014-16-1-2896, and N00014-20-1-2407. Funding Information: Parts of this work were funded by ONR grants N00014-18-1-2670, N00014-16-1-2896, and N00014-20-1-2407. Authors’ addresses: R. Liu and V. S. Subrahmanian, 6211 Sudikoff Lab, Dartmouth College, Hanover, NH 03755, USA; emails: {Rui.Liu.GR, vs}@dartmouth.edu; R. Liu, 1079 Commonwealth Avenue, APT 508, Boston, MA, 02215, USA; email: [email protected]; A. Pugliese, Cubo 44Z, University of Calabria, Rende, Italy; email: [email protected]. Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. Copyrights for components of this work owned by others than ACM must be honored. Abstracting with credit is permitted. To copy otherwise, or republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee. Request permissions from [email protected]. © 2020 Association for Computing Machinery. 2157-6904/2020/07-ART56 $15.00 https://doi.org/10.1145/3397463 Publisher Copyright: © 2020 ACM.

PY - 2020/9

Y1 - 2020/9

N2 - Customers of virtually all online marketplaces rely upon reviews in order to select the product or service they wish to buy. These marketplaces in turn deploy review fraud detection systems so that the integrity of reviews is preserved. A well-known problem with review fraud detection systems is their underlying assumption that the majority of reviews are honest-This assumption leads to a vulnerability where an attacker can try to generate many fake reviews of a product. In this article, we consider the case where a company wishes to fraudulently promote its product through fake reviews and propose the Sockpuppet-based Targeted Attack on Reviewing Systems (STARS for short). STARS enables an attacker to enter fake reviews for a product from multiple, apparently independent, sockpuppet accounts. We show that the STARS attack enables companies to successfully promote their product against seven recent, well-known review fraud detectors on four datasets (Amazon, Epinions, and the BitcoinAlpha and OTC exchanges) by significant margins. To protect against the STARS attack, we propose a new fraud detection algorithm called RTV. RTV introduces a new class of users (called trusted users) and also considers reviews left by verified users which were not considered in existing review fraud detectors. We show that RTV significantly mitigates the impact of the STARS attack across the four datasets listed above.

AB - Customers of virtually all online marketplaces rely upon reviews in order to select the product or service they wish to buy. These marketplaces in turn deploy review fraud detection systems so that the integrity of reviews is preserved. A well-known problem with review fraud detection systems is their underlying assumption that the majority of reviews are honest-This assumption leads to a vulnerability where an attacker can try to generate many fake reviews of a product. In this article, we consider the case where a company wishes to fraudulently promote its product through fake reviews and propose the Sockpuppet-based Targeted Attack on Reviewing Systems (STARS for short). STARS enables an attacker to enter fake reviews for a product from multiple, apparently independent, sockpuppet accounts. We show that the STARS attack enables companies to successfully promote their product against seven recent, well-known review fraud detectors on four datasets (Amazon, Epinions, and the BitcoinAlpha and OTC exchanges) by significant margins. To protect against the STARS attack, we propose a new fraud detection algorithm called RTV. RTV introduces a new class of users (called trusted users) and also considers reviews left by verified users which were not considered in existing review fraud detectors. We show that RTV significantly mitigates the impact of the STARS attack across the four datasets listed above.

KW - Data mining and knowledge discovery

KW - online commerce and recommendation systems

KW - social and information networks

UR - http://www.scopus.com/inward/record.url?scp=85091012105&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85091012105&partnerID=8YFLogxK

U2 - 10.1145/3397463

DO - 10.1145/3397463

M3 - Article

AN - SCOPUS:85091012105

SN - 2157-6904

VL - 11

JO - ACM Transactions on Intelligent Systems and Technology

JF - ACM Transactions on Intelligent Systems and Technology

IS - 5

M1 - 56

ER -

STARS: Defending against Sockpuppet-Based Targeted Attacks on Reviewing Systems

Abstract

Funding

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this