Sign In

Communications of the ACM

ACM TechNews

Say What? Scientists Devise an Algorithm That Detects Sarcasm Better Than Humans Can

View as: Print Mobile App Share:
The character Sheldon Cooper on the TV show "The Big Bang Theory" has trouble identifying sarcasm.

For the past decade, data scientists have been trying to develop algorithms that can automatically detect sarcasm.

Credit: Chuck Lorre Productions

Researchers at the University of California, Berkeley (UC Berkeley) and the University of Washington say they have developed a computer system that can identify the emotion behind a tweet with 85-percent accuracy, improving on previous research that recorded 75-percent accuracy.

The researchers were able to make the improvement by accounting for background information, such as details about the author, the audience to whom the tweet was directed, if it was a response, and the tweet to which it was responding.

The researchers focused on tweets containing the hashtags #sarcasm or #sarcastic.

The element most responsible for the increase in accuracy was information about the author. The researchers also found users with historically negative sentiments were more likely to be sarcastic. Although the contextual information produced a relatively small increase in accuracy, the researchers say the study highlights the importance of considering that information. "This gets into what is, at heart, so difficult about recognizing sarcasm--not just for computers, but for humans as well," says UC Berkeley professor David Bamman.

The researchers say future studies could refine the sarcasm detector to be even more accurate. For example, the researchers did not consider the fact that people are more likely to be sarcastic on some platforms than others.

From University of California, Berkeley
View Full Article


Abstracts Copyright © 2015 Information Inc., Bethesda, Maryland, USA


No entries found