Skip to Main content Skip to Navigation
Conference papers

You are what emojis say about your pictures: Language - independent gender inference attack on Facebook

Bizhan Alipour Pijani 1, 2 Abdessamad Imine 1, 2 Michaël Rusinowitch 1
1 PESTO - Proof techniques for security protocols
Inria Nancy - Grand Est, LORIA - FM - Department of Formal Methods
Abstract : The picture owner's gender has a strong influence on individuals' emotional reactions to the picture. In this study, we investigate gender inference attacks on their owners from pictures meta-data composed of: (i) alt-texts generated by Facebook to describe the content of pictures, and (ii) Emojis/Emoticons posted by friends, friends of friends or regular users as a reaction to the picture. Specifically, we study the correlation of picture owner gender with alt-text, and Emojis/Emoticons used by commenters when reacting to these pictures. We leverage this image sharing and reaction mode of Facebook users to derive an efficient and accurate technique for user gender inference. We show that such a privacy attack often succeeds even when other information than pictures published by their owners is either hidden or unavailable.
Complete list of metadatas

Cited literature [29 references]  Display  Hide  Download

https://hal.inria.fr/hal-02974078
Contributor : Michaël Rusinowitch <>
Submitted on : Wednesday, October 21, 2020 - 5:03:10 PM
Last modification on : Saturday, November 7, 2020 - 3:11:43 AM

File

sac.pdf
Files produced by the author(s)

Identifiers

Citation

Bizhan Alipour Pijani, Abdessamad Imine, Michaël Rusinowitch. You are what emojis say about your pictures: Language - independent gender inference attack on Facebook. SAC '20 - 35th ACM/SIGAPP Symposium on Applied Computing, Mar 2020, Brno, Czech Republic. pp.1826-1834, ⟨10.1145/3341105.3373943⟩. ⟨hal-02974078⟩

Share

Metrics

Record views

71

Files downloads

131