MNeagle
19th November 2012, 05:24 PM
http://i.huffpost.com/gen/864725/thumbs/r-TWITTER-HEAT-MAP-large570.jpg?5
Twitter may be full of jibber-jabber, but that doesn't mean this social networking site can't give us a little insight about what the world is thinking.
Silicon Graphics International (http://www.sgi.com/), or SGI, has partnered with researchers from the University of Illinois to scan international tweets in a project dubbed the Global Twitter Heartbeat (https://www.facebook.com/sgiglobal/app_164226463720371). By using SGI's UV 2000 Big Brain supercomputer (http://www.sgi.com/company_info/newsroom/press_releases/2012/june/uv.html), the researchers created real-time heat maps of positive and negative sentiments expressed via Twitter.
This means your angry/excited/nervous/happy tweets are possibly being recorded by an uber-speedy computer in order to stitch together an overall portrait of the Twittersphere's emotions.
So how does it work? "The project analyses every tweet to assign location (not just GPS-tagged tweets, but processing the text of the tweet itself), and tone values and then visualizes the conversation in a heat map infographic that combines and displays tweet location, intensity and tone," SGI's Facebook page reads (https://www.facebook.com/sgiglobal/app_164226463720371).
The Global Twitter Heartbeat tracks about 10 percent of the 500 million tweets posted daily -- that's approximately 50 million posts analyzed each day.
Thus far, SGI has created heat maps illustrating people's feelings on Twitter about Hurricane Sandy (https://www.youtube.com/watch?feature=player_embedded&v=g3AqdIDYG0c) and the 2012 election night (https://www.youtube.com/watch?feature=player_embedded&v=oVaBws-3BVs). To see the project in action, watch the clip below, showing how U.S. tweets were affected as the so-called "Frankenstorm (http://www.huffingtonpost.com/news/frankenstorm-2012)" barreled up the East Coast in late October. (Red patches represent negative sentiments. Blue patches are positive.)
http://www.youtube.com/watch?v=g3AqdIDYG0c&feature=player_embedded
Gigaom notes (http://gigaom.com/data/the-world-according-to-twitter-as-seen-through-a-high-performance-computer/) the UV 2 supercomputer that's used to process all of this emotion-filled Twitter data is an "impressive machine with a maximum of 4,096 cores." A press release from SGI (http://www.sgi.com/company_info/newsroom/press_releases/2012/june/uv.html) also states that the UV 2 could "ingest the entire contents of the U.S. Library of Congress print collection in less than three seconds."
http://www.huffingtonpost.com/2012/11/19/sgi-twitter-heat-map_n_2138726.html
Twitter may be full of jibber-jabber, but that doesn't mean this social networking site can't give us a little insight about what the world is thinking.
Silicon Graphics International (http://www.sgi.com/), or SGI, has partnered with researchers from the University of Illinois to scan international tweets in a project dubbed the Global Twitter Heartbeat (https://www.facebook.com/sgiglobal/app_164226463720371). By using SGI's UV 2000 Big Brain supercomputer (http://www.sgi.com/company_info/newsroom/press_releases/2012/june/uv.html), the researchers created real-time heat maps of positive and negative sentiments expressed via Twitter.
This means your angry/excited/nervous/happy tweets are possibly being recorded by an uber-speedy computer in order to stitch together an overall portrait of the Twittersphere's emotions.
So how does it work? "The project analyses every tweet to assign location (not just GPS-tagged tweets, but processing the text of the tweet itself), and tone values and then visualizes the conversation in a heat map infographic that combines and displays tweet location, intensity and tone," SGI's Facebook page reads (https://www.facebook.com/sgiglobal/app_164226463720371).
The Global Twitter Heartbeat tracks about 10 percent of the 500 million tweets posted daily -- that's approximately 50 million posts analyzed each day.
Thus far, SGI has created heat maps illustrating people's feelings on Twitter about Hurricane Sandy (https://www.youtube.com/watch?feature=player_embedded&v=g3AqdIDYG0c) and the 2012 election night (https://www.youtube.com/watch?feature=player_embedded&v=oVaBws-3BVs). To see the project in action, watch the clip below, showing how U.S. tweets were affected as the so-called "Frankenstorm (http://www.huffingtonpost.com/news/frankenstorm-2012)" barreled up the East Coast in late October. (Red patches represent negative sentiments. Blue patches are positive.)
http://www.youtube.com/watch?v=g3AqdIDYG0c&feature=player_embedded
Gigaom notes (http://gigaom.com/data/the-world-according-to-twitter-as-seen-through-a-high-performance-computer/) the UV 2 supercomputer that's used to process all of this emotion-filled Twitter data is an "impressive machine with a maximum of 4,096 cores." A press release from SGI (http://www.sgi.com/company_info/newsroom/press_releases/2012/june/uv.html) also states that the UV 2 could "ingest the entire contents of the U.S. Library of Congress print collection in less than three seconds."
http://www.huffingtonpost.com/2012/11/19/sgi-twitter-heat-map_n_2138726.html