Minimizing trust leaks for robust Sybil detection

János Höner, Shinichi Nakajima, Alexander Bauer, Klaus Muller, Nico Görnitz

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Sybil detection is a crucial task to protect online social networks (OSNs) against intruders who try to manipulate automatic services provided by OSNs to their customers. In this paper, we first discuss the robustness of graph-based Sybil detectors SybilRank and Integro and refine theoretically their security guarantees towards more realistic assumptions. After that, we formally introduce adversarial settings for the graph-based Sybil detection problem and derive a corresponding optimal attacking strategy by exploitation of trust leaks. Based on our analysis, we propose transductive Sybil ranking (TSR), a robust extension to SybilRank and Integro that directly minimizes trust leaks. Our empirical evaluation shows significant advantages of TSR over state-of-the-art competitors on a variety of attacking scenarios on artificially generated data and real-world datasets.

Original languageEnglish
Title of host publication34th International Conference on Machine Learning, ICML 2017
PublisherInternational Machine Learning Society (IMLS)
Pages2391-2401
Number of pages11
Volume4
ISBN (Electronic)9781510855144
Publication statusPublished - 2017 Jan 1
Event34th International Conference on Machine Learning, ICML 2017 - Sydney, Australia
Duration: 2017 Aug 62017 Aug 11

Other

Other34th International Conference on Machine Learning, ICML 2017
CountryAustralia
CitySydney
Period17/8/617/8/11

Fingerprint

Detectors

ASJC Scopus subject areas

  • Computational Theory and Mathematics
  • Human-Computer Interaction
  • Software

Cite this

Höner, J., Nakajima, S., Bauer, A., Muller, K., & Görnitz, N. (2017). Minimizing trust leaks for robust Sybil detection. In 34th International Conference on Machine Learning, ICML 2017 (Vol. 4, pp. 2391-2401). International Machine Learning Society (IMLS).

Minimizing trust leaks for robust Sybil detection. / Höner, János; Nakajima, Shinichi; Bauer, Alexander; Muller, Klaus; Görnitz, Nico.

34th International Conference on Machine Learning, ICML 2017. Vol. 4 International Machine Learning Society (IMLS), 2017. p. 2391-2401.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Höner, J, Nakajima, S, Bauer, A, Muller, K & Görnitz, N 2017, Minimizing trust leaks for robust Sybil detection. in 34th International Conference on Machine Learning, ICML 2017. vol. 4, International Machine Learning Society (IMLS), pp. 2391-2401, 34th International Conference on Machine Learning, ICML 2017, Sydney, Australia, 17/8/6.
Höner J, Nakajima S, Bauer A, Muller K, Görnitz N. Minimizing trust leaks for robust Sybil detection. In 34th International Conference on Machine Learning, ICML 2017. Vol. 4. International Machine Learning Society (IMLS). 2017. p. 2391-2401
Höner, János ; Nakajima, Shinichi ; Bauer, Alexander ; Muller, Klaus ; Görnitz, Nico. / Minimizing trust leaks for robust Sybil detection. 34th International Conference on Machine Learning, ICML 2017. Vol. 4 International Machine Learning Society (IMLS), 2017. pp. 2391-2401
@inproceedings{40fa14c7473b479fbb6b89ad1dcc5cb6,
title = "Minimizing trust leaks for robust Sybil detection",
abstract = "Sybil detection is a crucial task to protect online social networks (OSNs) against intruders who try to manipulate automatic services provided by OSNs to their customers. In this paper, we first discuss the robustness of graph-based Sybil detectors SybilRank and Integro and refine theoretically their security guarantees towards more realistic assumptions. After that, we formally introduce adversarial settings for the graph-based Sybil detection problem and derive a corresponding optimal attacking strategy by exploitation of trust leaks. Based on our analysis, we propose transductive Sybil ranking (TSR), a robust extension to SybilRank and Integro that directly minimizes trust leaks. Our empirical evaluation shows significant advantages of TSR over state-of-the-art competitors on a variety of attacking scenarios on artificially generated data and real-world datasets.",
author = "J{\'a}nos H{\"o}ner and Shinichi Nakajima and Alexander Bauer and Klaus Muller and Nico G{\"o}rnitz",
year = "2017",
month = "1",
day = "1",
language = "English",
volume = "4",
pages = "2391--2401",
booktitle = "34th International Conference on Machine Learning, ICML 2017",
publisher = "International Machine Learning Society (IMLS)",

}

TY - GEN

T1 - Minimizing trust leaks for robust Sybil detection

AU - Höner, János

AU - Nakajima, Shinichi

AU - Bauer, Alexander

AU - Muller, Klaus

AU - Görnitz, Nico

PY - 2017/1/1

Y1 - 2017/1/1

N2 - Sybil detection is a crucial task to protect online social networks (OSNs) against intruders who try to manipulate automatic services provided by OSNs to their customers. In this paper, we first discuss the robustness of graph-based Sybil detectors SybilRank and Integro and refine theoretically their security guarantees towards more realistic assumptions. After that, we formally introduce adversarial settings for the graph-based Sybil detection problem and derive a corresponding optimal attacking strategy by exploitation of trust leaks. Based on our analysis, we propose transductive Sybil ranking (TSR), a robust extension to SybilRank and Integro that directly minimizes trust leaks. Our empirical evaluation shows significant advantages of TSR over state-of-the-art competitors on a variety of attacking scenarios on artificially generated data and real-world datasets.

AB - Sybil detection is a crucial task to protect online social networks (OSNs) against intruders who try to manipulate automatic services provided by OSNs to their customers. In this paper, we first discuss the robustness of graph-based Sybil detectors SybilRank and Integro and refine theoretically their security guarantees towards more realistic assumptions. After that, we formally introduce adversarial settings for the graph-based Sybil detection problem and derive a corresponding optimal attacking strategy by exploitation of trust leaks. Based on our analysis, we propose transductive Sybil ranking (TSR), a robust extension to SybilRank and Integro that directly minimizes trust leaks. Our empirical evaluation shows significant advantages of TSR over state-of-the-art competitors on a variety of attacking scenarios on artificially generated data and real-world datasets.

UR - http://www.scopus.com/inward/record.url?scp=85048393227&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85048393227&partnerID=8YFLogxK

M3 - Conference contribution

VL - 4

SP - 2391

EP - 2401

BT - 34th International Conference on Machine Learning, ICML 2017

PB - International Machine Learning Society (IMLS)

ER -