Back To Schedule
Thursday, November 19 • 17:15 - 18:05
Expression: Understanding Affective Content in Social Media

Sign up or log in to save this to your schedule, view media, leave feedback and see who's attending!

Analysis of social content for understanding people's sentiments towards topics of interest that change over time has become an attractive and challenging research area. Natural Language Processing (NLP) techniques are being adapted to deal with streams of social content. New visualization approaches need also to be proposed to express, in a user friendly and reactive manner, individual as well as collective sentiments. In this paper, we present Expression, an integrated framework that allows users to express their opinions through a social platform and to see others' comments. We introduce the Sentiment Card concept: a live representation of a topic of interest. The Sentiment Card is a space that allows users to express their comments and to understand the trend of selected topics of interest expressed by other users. The design of Expression is presented, describing in particular, the sentiment classification module as well as the sentiment card visualization component. Results of the evaluation of our prototype by a usability study are also discussed and considered for motivating future research.

Thursday November 19, 2015 17:15 - 18:05 GMT-03
Aula 4.6 Rondeau 165

Attendees (0)