Abstract
Profile images on social networks are users’ opportunity topresent themselves and to affect how others judge them. Weexamine what Facebook images say about users’ perceivedand measured intelligence. 1,122 Facebook users completeda matrices intelligence test and shared their current Facebookprofile image. Strangers also rated the images for perceivedintelligence. We use automatically extracted imagefeatures to predict both measured and perceived intelligence.Intelligence estimation from images is a difficult task evenfor humans, but experimental results show that human accuracycan be equalled using computing methods. We reportthe image features that predict both measured and perceivedintelligence, and highlight misleading features suchas "smiling” and "wearing glasses” that are correlated withperceived but not measured intelligence. Our results give insightsinto inaccurate stereotyping from profile images andalso have implications for privacy, especially since in mostsocial networks profile images are public by default.
Original language | English |
---|---|
Title of host publication | Proceedings of the 10th ACM international conference on Web search and data mining |
Place of Publication | New York, U. S. A. |
Publisher | Association for Computing Machinery |
Pages | 33-40 |
Number of pages | 8 |
ISBN (Print) | 9781450346757 |
DOIs | |
Publication status | Published - 6 Feb 2017 |
Keywords
- Intelligence quotient
- Perceived intelligence
- Intelligence estimation
- Computational aesthetics
- IQ
- Measured intelligence