Looking Good With Flickr Faves: Gaussian Processes for Finding Difference Makers in Personality Impressions

Xiong, X., Filippone, M. and Vinciarelli, A. (2016) Looking Good With Flickr Faves: Gaussian Processes for Finding Difference Makers in Personality Impressions. In: 2016 ACM Multimedia Conference (MM '16), Amsterdam, The Netherlands, 15-19 Oct 2016, pp. 412-415. ISBN 9781450336031 (doi:10.1145/2964284.2967253)

[img]
Preview
Text
145764.pdf - Accepted Version

447kB

Abstract

Flickr allows its users to generate galleries of "faves", i.e., pictures that they have tagged as favourite. According to recent studies, the faves are predictive of the personality traits that people attribute to Flickr users. This article investigates the phenomenon and shows that faves allow one to predict whether a Flickr user is perceived to be above median or not with respect to each of the Big-Five Traits (accuracy up to 79\% depending on the trait). The classifier - based on Gaussian Processes with a new kernel designed for this work - allows one to identify the visual characteristics of faves that better account for the prediction outcome.

Item Type:Conference Proceedings
Status:Published
Refereed:Yes
Glasgow Author(s) Enlighten ID:Xiong, Xiaoyu and Filippone, Dr Maurizio and Vinciarelli, Professor Alessandro
Authors: Xiong, X., Filippone, M., and Vinciarelli, A.
College/School:College of Science and Engineering > School of Computing Science
ISBN:9781450336031
Related URLs:

University Staff: Request a correction | Enlighten Editors: Update this record