Exploiting Innocuous Activity for Correlating Users Across Sites

TitleExploiting Innocuous Activity for Correlating Users Across Sites
Publication TypeConference Paper
Year of Publication2013
AuthorsGoga, O., Lei H., Parthasarathi S. Hari Krish, Friedland G., Sommer R., & Teixeira R.
Other Numbers3411

We study how potential attackers can identify accounts on differentsocial network sites that all belong to the same user, exploitingonly innocuous activity that inherently comes with posted content.We examine three specific features on Yelp, Flickr, and Twitter: thegeo-location attached to a user’s posts, the timestamp of posts, andthe user’s writing style as captured by language models. We showthat among these three features the location of posts is the mostpowerful feature to identify accounts that belong to the same userin different sites. When we combine all three features, the accuracyof identifying Twitter accounts that belong to a set of Flickr usersis comparable to that of existing attacks that exploit usernames.


This work was supported by the National Science Foundation undergrant CNS-1065240, and by the Agence National de la Recherchegrant C’MON and was carried out at LINCS (www.lincs.fr)and ICSI (www.icsi.berkeley.edu). Any opinions, findings,and conclusions or recommendations expressed in this materialare those of the authors or originators and do not necessarily reflectthe views of the NSF or ANR. We would like to thank PatrickLoiseau for his valuable feedback during all stages of the project,as well as the anonymous reviewers for their helpful comments.

Bibliographic Notes

Proceedings of the World Wide Web Conference (WWW), Rio de Janeiro, Brazil

Abbreviated Authors

O. Goga, H. Lei, S. H. K. Parthasarathi, G. Friedland, R. Sommer, and R. Teixeira

ICSI Research Group

Audio and Multimedia

ICSI Publication Type

Article in conference proceedings