Dynamic surveillance: A case study with Enron email data set

Heesung Do, Peter Choi, Heejo Lee

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

1 Citation (Scopus)

Abstract

Surveillance is a critical measure to break anonymity. While surveillance with unlimited resources is often assumed as a means, against which, to design stronger anonymity algorithms, this paper addresses the general impact of limited resource on surveillance efficiency. The general impact of limited resource on identifying a hidden group is experimentally studied; the task of identification is only done by following communications between suspects, i.e., the information of whos talking to whom. The surveillance uses simple but intuitive algorithms to return more intelligence with limited resource. The surveillance subject used in this work is the publicly available Enron email data set, an actual trace of human interaction. The initial expectation was that, even with limited resource, intuitive surveillance algorithms would return the higher intelligence than a random approach by exploiting the general properties of power law-style communication map. To the contrary, the impact of limited resource was found large to the extent that intuitive algorithms do not return significantly higher intelligence than a random approach.

Original languageEnglish
Title of host publicationLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
PublisherSpringer Verlag
Pages81-99
Number of pages19
Volume8267 LNCS
ISBN (Print)9783319051482
DOIs
Publication statusPublished - 2014
Event14th International Workshop on Information Security Applications, WISA 2013 - Jeju Island, Korea, Republic of
Duration: 2013 Aug 192013 Aug 21

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume8267 LNCS
ISSN (Print)03029743
ISSN (Electronic)16113349

Other

Other14th International Workshop on Information Security Applications, WISA 2013
CountryKorea, Republic of
CityJeju Island
Period13/8/1913/8/21

Fingerprint

Electronic mail
Electronic Mail
Surveillance
Resources
Intuitive
Anonymity
Communication
Power Law
Trace
Interaction
Intelligence

Keywords

  • Anonymity
  • Budget
  • Email data set
  • Surveillanc

ASJC Scopus subject areas

  • Computer Science(all)
  • Theoretical Computer Science

Cite this

Do, H., Choi, P., & Lee, H. (2014). Dynamic surveillance: A case study with Enron email data set. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 8267 LNCS, pp. 81-99). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 8267 LNCS). Springer Verlag. https://doi.org/10.1007/978-3-319-05149-9_6

Dynamic surveillance : A case study with Enron email data set. / Do, Heesung; Choi, Peter; Lee, Heejo.

Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 8267 LNCS Springer Verlag, 2014. p. 81-99 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 8267 LNCS).

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

Do, H, Choi, P & Lee, H 2014, Dynamic surveillance: A case study with Enron email data set. in Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). vol. 8267 LNCS, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 8267 LNCS, Springer Verlag, pp. 81-99, 14th International Workshop on Information Security Applications, WISA 2013, Jeju Island, Korea, Republic of, 13/8/19. https://doi.org/10.1007/978-3-319-05149-9_6
Do H, Choi P, Lee H. Dynamic surveillance: A case study with Enron email data set. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 8267 LNCS. Springer Verlag. 2014. p. 81-99. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). https://doi.org/10.1007/978-3-319-05149-9_6
Do, Heesung ; Choi, Peter ; Lee, Heejo. / Dynamic surveillance : A case study with Enron email data set. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 8267 LNCS Springer Verlag, 2014. pp. 81-99 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
@inproceedings{bd9c970113114c7788e0d336b0ff6dee,
title = "Dynamic surveillance: A case study with Enron email data set",
abstract = "Surveillance is a critical measure to break anonymity. While surveillance with unlimited resources is often assumed as a means, against which, to design stronger anonymity algorithms, this paper addresses the general impact of limited resource on surveillance efficiency. The general impact of limited resource on identifying a hidden group is experimentally studied; the task of identification is only done by following communications between suspects, i.e., the information of whos talking to whom. The surveillance uses simple but intuitive algorithms to return more intelligence with limited resource. The surveillance subject used in this work is the publicly available Enron email data set, an actual trace of human interaction. The initial expectation was that, even with limited resource, intuitive surveillance algorithms would return the higher intelligence than a random approach by exploiting the general properties of power law-style communication map. To the contrary, the impact of limited resource was found large to the extent that intuitive algorithms do not return significantly higher intelligence than a random approach.",
keywords = "Anonymity, Budget, Email data set, Surveillanc",
author = "Heesung Do and Peter Choi and Heejo Lee",
year = "2014",
doi = "10.1007/978-3-319-05149-9_6",
language = "English",
isbn = "9783319051482",
volume = "8267 LNCS",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
publisher = "Springer Verlag",
pages = "81--99",
booktitle = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

}

TY - GEN

T1 - Dynamic surveillance

T2 - A case study with Enron email data set

AU - Do, Heesung

AU - Choi, Peter

AU - Lee, Heejo

PY - 2014

Y1 - 2014

N2 - Surveillance is a critical measure to break anonymity. While surveillance with unlimited resources is often assumed as a means, against which, to design stronger anonymity algorithms, this paper addresses the general impact of limited resource on surveillance efficiency. The general impact of limited resource on identifying a hidden group is experimentally studied; the task of identification is only done by following communications between suspects, i.e., the information of whos talking to whom. The surveillance uses simple but intuitive algorithms to return more intelligence with limited resource. The surveillance subject used in this work is the publicly available Enron email data set, an actual trace of human interaction. The initial expectation was that, even with limited resource, intuitive surveillance algorithms would return the higher intelligence than a random approach by exploiting the general properties of power law-style communication map. To the contrary, the impact of limited resource was found large to the extent that intuitive algorithms do not return significantly higher intelligence than a random approach.

AB - Surveillance is a critical measure to break anonymity. While surveillance with unlimited resources is often assumed as a means, against which, to design stronger anonymity algorithms, this paper addresses the general impact of limited resource on surveillance efficiency. The general impact of limited resource on identifying a hidden group is experimentally studied; the task of identification is only done by following communications between suspects, i.e., the information of whos talking to whom. The surveillance uses simple but intuitive algorithms to return more intelligence with limited resource. The surveillance subject used in this work is the publicly available Enron email data set, an actual trace of human interaction. The initial expectation was that, even with limited resource, intuitive surveillance algorithms would return the higher intelligence than a random approach by exploiting the general properties of power law-style communication map. To the contrary, the impact of limited resource was found large to the extent that intuitive algorithms do not return significantly higher intelligence than a random approach.

KW - Anonymity

KW - Budget

KW - Email data set

KW - Surveillanc

UR - http://www.scopus.com/inward/record.url?scp=84958533915&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84958533915&partnerID=8YFLogxK

U2 - 10.1007/978-3-319-05149-9_6

DO - 10.1007/978-3-319-05149-9_6

M3 - Conference contribution

AN - SCOPUS:84958533915

SN - 9783319051482

VL - 8267 LNCS

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 81

EP - 99

BT - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

PB - Springer Verlag

ER -