BEARS 2015 Preview: Multimedia Opinion Mining with Visual Sentiment Analysis

Thursday, February 5, 2015

 

Damian Borth

ICSI's 2015 open house is Thursday, February 12 at 12:45 p.m., and to get ready for it , we're previewing some of the 16 demos and posters our researchers will be presenting.

Damian Borth is a postdoc in the Audio and Multimedia Group, visiting us on a DAAD-funded postdoc. In his past work, he's looked at visual sentiment analysis, a useful tool for analyzing social multimedia posts - those that incorporate, say, text and images - since text often doesn't convey the whole meaning of a post. Read more about his work here. At the open house, he'll be presenting his new work on multimedia opinion mining.

Learn more about ICSI's 2015 BEARS Open House.


Multimedia Opinion Mining with Visual Sentiment Analysis

The use of images and videos in social media is increasing and gives rise to a new type of content called "social multimedia." Such content conveys much about our thinking, in that it reflects our personal values and ourselves as a society. To make this content accessible, we present SentiBank, a framework for visual sentiment analysis consisting of 1,500 adjective-noun pair (ANP) classifiers trained from Caffe's deep learning features. Further, we would like to show applications around its usage for social good or brand monitoring.

 

a screenshot of the Visual Sentiment Ontology treemap

a screenshot of the Visual Sentiment Ontology

 

Add new comment

Plain text

  • No HTML tags allowed.
  • Web page addresses and e-mail addresses turn into links automatically.
  • Lines and paragraphs break automatically.