Popular Science reports that a team from the Computer Science Department at The Hebrew University in Jerusalem, Israel, has come up with technical – and subtle – solution for identifying sarcasm in electronic text.
Oren Tsur, Dmitry Davidov and Ari Rappoport have devised an algorithm that identifies sarcastic tweets on Twitter and sarcastic comments in product reviews on Amazon. Named SASI, or Semi-supervised Algorithm for Sarcasm Identification, the formula successfully identified such sentences as: “Great idea, now try again with a real product development team”.This could be even more valuable than Snopes, so that sarcasm-challenged readers and commenters can find out in advance if what they are reading or responding to is meant to be sarcastic and cynical.
In a paper titled Semi-Supervised Recognition of Sarcastic Sentences in Twitter and Amazon, the team claims the sarcasm detection algorithm could have important benefits such as personalising content ranking and recommendation systems such as those favoured by Amazon. Since sarcastic comment involves stating the opposite of what you actually mean, it could also be useful for repairing inverse allocations in opinion mining systems – and presumably pave the way for sarcasm-sensitive robots. The team’s findings will be presented at the International Conference for Weblogs and Social Media in Washington, DC, next week.
It probably wouldn't work on "Far Side" comics though.
One-Click Pro-Israel Activism!
It takes just 1 Click to help Israel!
Wherever I am, my blog turns towards Eretz Yisrael טובה הארץ מאד מאד
