Rating is available when the video has been rented.
This feature is not available right now. Please try again later.
Published on Oct 5, 2013
This video demonstrates automatic sentiment tracking of image tweets using visual content analysis. The dataset includes 2,000 image tweets collected during 2012 Hurricane Sandy. We use SentiBank, consisting of 1,200 visual sentiment classifiers, to extract visual concepts and predict the sentiment expressed in the pictures. The results were compared with predictions obtained by tweet text analysis (SentiStrength) and text-image combination. The text-visual combined approach reaches the highest accuracy at 72%.
paper: Damian Borth, Rongrong Ji, Tao Chen, Thomas Breuel and Shih-Fu Chang, "Large-scale Visual Sentiment Ontology and Detectors Using Adjective Noun Pairs," ACM Int. Conference on Multimedia (ACM MM), Barcelona, Oct 2013. (video produced by Tao Chen)