Analyzing and Predicting Sentiment of Images on the Social Web
Analyzing and Predicting Sentiment of Images on the Social Web
In this paper we study the connection between sentiment of images expressed in metadata and their visual content in the social photo sharing environment Flickr. To this end, we consider the bag-of-visual words representation as well as the color distribution of images, and make use of the SentiWordNet thesaurus to extract numerical values for their sentiment from accompanying textual metadata. We then perform a discriminative feature analysis based on information theoretic methods, and apply machine learning techniques to predict the sentiment of images. Our large-scale empirical study on a set of over half a million Flickr images shows a considerable correlation between sentiment and visual features, and promising results towards estimating the polarity of sentiment in images.
Color features, Classification, Sentiment Analysis, Automatic Annotation, Visual Terms
978-1-60558-933-6
715-718
Siersdorfer, Stefan
e5a8a356-ceeb-4d95-b684-98992e549714
Hare, Jonathon
65ba2cda-eaaf-4767-a325-cd845504e5a9
Minack, Enrico
db06f699-4d0e-4767-b7c8-05bcec8990a1
Deng, Fan
1a90c639-73a2-400d-9ed7-08d4b26a7375
25 October 2010
Siersdorfer, Stefan
e5a8a356-ceeb-4d95-b684-98992e549714
Hare, Jonathon
65ba2cda-eaaf-4767-a325-cd845504e5a9
Minack, Enrico
db06f699-4d0e-4767-b7c8-05bcec8990a1
Deng, Fan
1a90c639-73a2-400d-9ed7-08d4b26a7375
Siersdorfer, Stefan, Hare, Jonathon, Minack, Enrico and Deng, Fan
(2010)
Analyzing and Predicting Sentiment of Images on the Social Web.
ACM Multimedia 2010, Firenze, Italy.
25 - 29 Oct 2010.
.
Record type:
Conference or Workshop Item
(Paper)
Abstract
In this paper we study the connection between sentiment of images expressed in metadata and their visual content in the social photo sharing environment Flickr. To this end, we consider the bag-of-visual words representation as well as the color distribution of images, and make use of the SentiWordNet thesaurus to extract numerical values for their sentiment from accompanying textual metadata. We then perform a discriminative feature analysis based on information theoretic methods, and apply machine learning techniques to predict the sentiment of images. Our large-scale empirical study on a set of over half a million Flickr images shows a considerable correlation between sentiment and visual features, and promising results towards estimating the polarity of sentiment in images.
Text
p715.pdf
- Version of Record
More information
Published date: 25 October 2010
Additional Information:
Event Dates: 25-29 October 2010
Venue - Dates:
ACM Multimedia 2010, Firenze, Italy, 2010-10-25 - 2010-10-29
Keywords:
Color features, Classification, Sentiment Analysis, Automatic Annotation, Visual Terms
Organisations:
Web & Internet Science
Identifiers
Local EPrints ID: 271670
URI: http://eprints.soton.ac.uk/id/eprint/271670
ISBN: 978-1-60558-933-6
PURE UUID: 5d74f73a-10df-4562-ae70-8656e2eb975d
Catalogue record
Date deposited: 02 Nov 2010 17:59
Last modified: 15 Mar 2024 03:25
Export record
Contributors
Author:
Stefan Siersdorfer
Author:
Jonathon Hare
Author:
Enrico Minack
Author:
Fan Deng
Download statistics
Downloads from ePrints over the past year. Other digital versions may also be available to download e.g. from the publisher's website.
View more statistics