November 24, 2024
Researchers say they have trained an artificial intelligence model to be able to detect sarcasm, according to a Thursday report. The research was presented at the Acoustical Society of America and the Canadian Acoustical Association in Ottawa by Netherlands-based researchers, according to the Guardian. Researchers said they trained the AI model using scenes from popular […]

Researchers say they have trained an artificial intelligence model to be able to detect sarcasm, according to a Thursday report.

The research was presented at the Acoustical Society of America and the Canadian Acoustical Association in Ottawa by Netherlands-based researchers, according to the Guardian.

Researchers said they trained the AI model using scenes from popular TV shows, focusing on Chandler Bing from Friends and Sheldon Cooper from The Big Bang Theory. The data included Chandler responding to the idea of spending a day putting furniture together with, “Yes, and we’re very excited about it,” and Sheldon telling another character, “It’s just a privilege to watch your mind at work,” according to the report.

“We are able to recognize sarcasm in a reliable way, and we’re eager to grow that,” Matt Coler at the University of Groningen’s speech technology lab told the outlet. “We want to see how far we can push it.”

Researchers said multiple cues were necessary when effectively attempting to identify sarcasm, that text alone would be more difficult for a sarcasm detector to decipher, and that communicating human emotions to an AI model provided difficulties.

“When you start studying sarcasm, you become hyper-aware of the extent to which we use it as part of our normal mode of communication,” Coler said. “But we have to speak to our devices in a very literal way, as if we’re talking to a robot, because we are. It doesn’t have to be this way.”

[embedded content]

Researchers said their model could detect sarcasm in sitcoms approximately 75% of the time, though the exact methodology or what is considered a positive identification of sarcasm was not laid out in the report.

CLICK HERE TO READ MORE FROM THE WASHINGTON EXAMINER

The researchers acknowledged the difficulty in identifying sarcasm, even between two humans.

“Are we going to have a machine that is 100% accurate?” said Xiyuan Gao, a Ph.D. student at the lab. “That’s not something even humans can achieve.”

Leave a Reply