Stylometric linkability of tweets

Mishari Almishari; Mohamed Ali Kaafar; Ekin Oguz; Gene Tsudik

doi:10.1145/2665943.2665966

Stylometric linkability of tweets

Mishari Almishari^*, Mohamed Ali Kaafar, Ekin Oguz, Gene Tsudik

^*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceeding › Conference proceeding contribution › peer-review

10 Citations (Scopus)

Abstract

Microblogging is a very popular Internet activity that informs and entertains a large number of people via terse messages; e.g., tweets on Twitter. Even though microblogging does not emphasize privacy, authors can easily hide behind pseudonyms and multiple accounts on the same, or across multiple, site(s). In this paper, we explore stylometric linkability of tweets. Our results clearly demonstrate that multiple sets of tweets by the same author are easily linkable even when the number of possible authors is large. This is also confirmed by showing that linkability holds for a set of actual Twitter users who admittedly tweet via multiple accounts.

Original language	English
Title of host publication	Proceedings of the 13th Workshop on Privacy in the Electronic Society
Place of Publication	New York
Publisher	Association for Computing Machinery
Pages	205-208
Number of pages	4
ISBN (Electronic)	9781450331487
DOIs	https://doi.org/10.1145/2665943.2665966
Publication status	Published - 3 Nov 2014
Externally published	Yes
Event	13th Workshop on Privacy in the Electronic Society, WPES 2014, in Conjunction with the ACM Conference on Computer and Communications Security, ACM CCS 2014 - Scottsdale, United States Duration: 3 Nov 2014 → …

Conference

Conference	13th Workshop on Privacy in the Electronic Society, WPES 2014, in Conjunction with the ACM Conference on Computer and Communications Security, ACM CCS 2014
Country/Territory	United States
City	Scottsdale
Period	3/11/14 → …

Keywords

author attribution
linkability
microblogging

Access to Document

10.1145/2665943.2665966

Cite this

@inproceedings{4ffbdcba76d64c0b945079e5586a3a39,

title = "Stylometric linkability of tweets",

abstract = "Microblogging is a very popular Internet activity that informs and entertains a large number of people via terse messages; e.g., tweets on Twitter. Even though microblogging does not emphasize privacy, authors can easily hide behind pseudonyms and multiple accounts on the same, or across multiple, site(s). In this paper, we explore stylometric linkability of tweets. Our results clearly demonstrate that multiple sets of tweets by the same author are easily linkable even when the number of possible authors is large. This is also confirmed by showing that linkability holds for a set of actual Twitter users who admittedly tweet via multiple accounts.",

keywords = "author attribution, linkability, microblogging",

author = "Mishari Almishari and Kaafar, {Mohamed Ali} and Ekin Oguz and Gene Tsudik",

year = "2014",

month = nov,

day = "3",

doi = "10.1145/2665943.2665966",

language = "English",

pages = "205--208",

booktitle = "Proceedings of the 13th Workshop on Privacy in the Electronic Society",

publisher = "Association for Computing Machinery",

note = "13th Workshop on Privacy in the Electronic Society, WPES 2014, in Conjunction with the ACM Conference on Computer and Communications Security, ACM CCS 2014 ; Conference date: 03-11-2014",

}

Almishari, M, Kaafar, MA, Oguz, E & Tsudik, G 2014, Stylometric linkability of tweets. in Proceedings of the 13th Workshop on Privacy in the Electronic Society. Association for Computing Machinery, New York, pp. 205-208, 13th Workshop on Privacy in the Electronic Society, WPES 2014, in Conjunction with the ACM Conference on Computer and Communications Security, ACM CCS 2014, Scottsdale, United States, 3/11/14. https://doi.org/10.1145/2665943.2665966

TY - GEN

T1 - Stylometric linkability of tweets

AU - Almishari, Mishari

AU - Kaafar, Mohamed Ali

AU - Oguz, Ekin

AU - Tsudik, Gene

PY - 2014/11/3

Y1 - 2014/11/3

N2 - Microblogging is a very popular Internet activity that informs and entertains a large number of people via terse messages; e.g., tweets on Twitter. Even though microblogging does not emphasize privacy, authors can easily hide behind pseudonyms and multiple accounts on the same, or across multiple, site(s). In this paper, we explore stylometric linkability of tweets. Our results clearly demonstrate that multiple sets of tweets by the same author are easily linkable even when the number of possible authors is large. This is also confirmed by showing that linkability holds for a set of actual Twitter users who admittedly tweet via multiple accounts.

AB - Microblogging is a very popular Internet activity that informs and entertains a large number of people via terse messages; e.g., tweets on Twitter. Even though microblogging does not emphasize privacy, authors can easily hide behind pseudonyms and multiple accounts on the same, or across multiple, site(s). In this paper, we explore stylometric linkability of tweets. Our results clearly demonstrate that multiple sets of tweets by the same author are easily linkable even when the number of possible authors is large. This is also confirmed by showing that linkability holds for a set of actual Twitter users who admittedly tweet via multiple accounts.

KW - author attribution

KW - linkability

KW - microblogging

UR - http://www.scopus.com/inward/record.url?scp=84910634883&partnerID=8YFLogxK

U2 - 10.1145/2665943.2665966

DO - 10.1145/2665943.2665966

M3 - Conference proceeding contribution

AN - SCOPUS:84910634883

SP - 205

EP - 208

BT - Proceedings of the 13th Workshop on Privacy in the Electronic Society

PB - Association for Computing Machinery

CY - New York

T2 - 13th Workshop on Privacy in the Electronic Society, WPES 2014, in Conjunction with the ACM Conference on Computer and Communications Security, ACM CCS 2014

Y2 - 3 November 2014

ER -

Stylometric linkability of tweets

Abstract

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this