Computer science researchers together with one in all Indian-origin on the University of Central Florida have developed a man-made intelligence (AI)-based sarcasm detector for posts on social media platforms. Sarcasm has been a significant hurdle to rising the accuracy of sentiment evaluation, particularly on social media, since sarcasm depends closely on vocal tones, facial expressions and gestures that can’t be represented in the textual content.
While synthetic intelligence (AI) refers to logical knowledge evaluation and response, sentiment evaluation is akin to accurately figuring out emotional communication on social media. “The presence of sarcasm in the text is the main hindrance in the performance of sentiment analysis,” says Ivan Garibay, Assistant Professor of engineering from Complex Adaptive Systems Lab (CASL) on the University of Central Florida.
Sarcasm isn’t at all times straightforward to determine in dialog, so that you can think about it’s fairly difficult for a pc programme to do it and do it effectively. “We developed an interpretable deep learning model using multi-head self-attention and gated recurrent units,” Garibay mentioned in a paper revealed in the journal Entropy.
The workforce taught the pc mannequin to seek out patterns that always point out sarcasm and mixed that with instructing the programme to accurately pick cue phrases in sequences that have been extra prone to point out sarcasm. They taught the mannequin to do that by feeding it giant knowledge units after which checked its accuracy.
The workforce included pc science doctoral scholar Ramya Akula. “In face-to-face conversation, sarcasm can be identified effortlessly using facial expressions, gestures, and tone of the speaker,” Akula mentioned.
“Detecting sarcasm in textual communication is not a trivial task as none of these cues are readily available. Especially with the explosion of internet usage, sarcasm detection in online communications from social networking platforms is much more challenging,” she added.
#mute
