Mapping Twitter Emotions

On October 7, 2010, in GIS DATA, OTHER, by Eric Edmonds

Viswa Mavi, an intern at Microsoft, created “TweetHeat” – a mapping application that geographically references Twitter posts and their emotional responses towards a subject.  This map shows how nobody in the Cleveland area has anything good to say about LeBron James (his first game with the Heat was this week).

“The map app helps users visualize the public response/feeling towards a product, event or really any topic. All tweets related to the searched keyword are analyzed using Natural Language Processing, more specifically Sentiment Analysis. The analysis assigns a score from 1-5, from a very negative to very positive emotion in the tweet. The analyzed tweets are mapped on Bing Maps based on where they were tweeted from.

“To help bring the visualization to life, I created two views. The first shows individual tweets on the map colored according to the emotion of the tweet. A red tweet indicates a score of ’1′, or very negative emotion. A green tweet indicates a score of ’5′, or very positive emotion. Other shades fall in between according to emotion. The second visualization provides you with the overall emotion by area. This view colors the map area to convey an overall twitter user feeling.”

Related posts:

  1. Interactive GIS Mapping Sites in TN
  2. What? So now we’re mapping termites?
  3. Full disclosure property mapping
  4. Southeast Watershed Forum Announces New Online Mapping Service
  5. SERUG Underway
Tagged with:  

Leave a Reply

Your email address will not be published. Required fields are marked *

*

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>