Researchers develop a method of detecting sarcasm in tweets that is better than humans at correctly identifying if a tweet is sarcastic.
They also tested it against humans, using volunteers recruited through the crowdsourcing site Mechanical Turk. They found it was better than the humans at spotting sarcasm and other emotions on Twitter. It was 82 percent accurate at identifying sarcasm correctly, compared with an average score of 76 percent for the human volunteers.
An algorithm trained on emoji knows when you’re being sarcastic on Twitter
Understanding sarcasm could help AI fight racism, abuse, and harassment.
now the real question is, is it worth to be sarcastic when people can't get it, or is this the only case where sarcasm actually works as intended?
How do they know if the data they tested against was being sarcastic enough? Surely at one point they must have asked a human who may or may not have been lying.
Also, just assuming that this isn't going to turn out to be super culturally specific.
This seems like a step towards recognising puns – vital for crossword solving?