The U.S.-Germany Match Through a Social Media Lens – NYTimes …
Nicholas Kamm/Agence France-Presse — Getty ImagesWorld Cup fans gathered in Washington on Thursday to watch the United States play Germany.
The United States-Germany soccer match on Thursday was a roller coaster of emotions. And an analysis of the social media communications before, during and after the match shows the minute-by-minute shifts in sentiment and in subjects discussed, as the American team lost to Germany but moved on in the World Cup competition.
Luminoso, a text analysis and artificial intelligence start-up, which grew out of research done at the M.I.T. Media Lab, created a visualization of the social media communications on Twitter, Facebook and Google Plus, 900,000 in all.
The graphic shows, first, the optimism before the match began, as reflected in the high volume of posts with the hashtag #IBelieveWeWillWin.
Then, reality sets in, as a strong German team dominates in time of possession and is the aggressor. We see three elements: messages with negative and positive sentiment, and more negative than positive; and discussion of the American team’s defense, which figures prominently since German has the ball most of the time. Two beleaguered American midfielders, Michael Bradley and Jermaine Jones, are the most criticized players.
Later in the match, discussion of the crucial game between Portugal and Ghana, being played at the same time in another stadium in Brazil, begins to bubble up. An assist from Portugal, which beat Ghana 1-0, ensured that the United States would move on in the World Cup. After the game, the online messages shift again to optimism — lots of #IBelieveWeWillWin posts — and to comments about the Ghana-Portugal match and its implications for the United States team.
Luminoso has been conducting this analysis and sorting throughout the World Cup, mainly to feed Sony’s One Stadium Live website and mobile application. One feature of this match, not seen during earlier contests, was the social-media cross talk — the Portugal-Ghana discussion during the United States-Germany match. “This time, you see conversations interacting,” said Catherine Havasi, co-founder and chief executive of Luminoso. “That is a different kind of dynamic than we had seen before.”
Luminoso’s software goes beyond sentiment analysis, which correlates words and phases with positive or negative emotions. “That was great.” “That was horrible.” “Go team.” Its technology also includes a knowledge base, which is built off a crowdsourced project at the Massachusetts Institute of Technology that began in 1999 and is called the Open Mind Common Sense Project.
So its smart technology understands, in its way, concepts instead of just words. Examples of such concepts, said Ms. Havasi, a research scientist in artificial intelligence and computational linguistics, are: “The sun is hot.” “Coffee comes in mugs.” “People want to be respected.”
It uses those techniques to tease out soccer concepts that can be inferred, if not stated directly, like “US Defense” in the graphic above. “We do that in a fairly unstructured way, so it can be done as the data streams by, on the fly,” Ms. Havasi said.
Luminoso, founded in 2010, mostly works with consumer product companies and retailers, including Pepsi and Amazon.
Jump to original:
The U.S.-Germany Match Through a Social Media Lens – NYTimes …
