Researchers develop AI model able to detect sarcasm in social media
Researches at the University of Central Florida, USA, have developed an artificial intelligence (AI) model that can detect sarcasm in social media posts. The team trained the computer model – by using large sets of data and repeatedly checking its accuracy – to (a) find patterns that often indicate sarcasm and (b) correctly pick out cue words in sequences that were more likely to indicate sarcasm. As Ivan Garibay, Assistant Professor of engineering, explains, the researchers ‘developed an interpretable deep learning model using multi-head self-attention and gated recurrent units. The multi-head self-attention module aids in identifying crucial sarcastic cue-words from the input, and the recurrent units learn long-range dependencies between these cue-words to better classify the input text.’ The work was carried out in the framework of a grant offered by the US Defense Advanced Research Project Agency (DARPA) in support of its Computational Simulation of Online Social Behaviour programme.