Researchers at the University of Central Florida have developed a new artificial intelligence (AI) tool that is capable of detecting sarcasm in social media. According to the team, this type of tool is highly useful for companies looking to better understand and respond to customer feedback on top social media platforms like Twitter and Facebook. It is extremely difficult to keep up with this process manually.
One of the major aspects of the tool is sentiment analysis, which is the automated process of identifying positive, negative, and neutral emotions within text. Sentiment analysis is focused on identifying emotional communication, while AI is focused on logical analysis and response.
The new research was published in the journal Entropy.
Teaching the Model to Detect Sarcasm
The computer model was taught to detect patterns that indicate sarcasm, and it was taught to identify specific cue words in a sentence that indicated sarcasm. This was accomplished by the team feeding the model large data sets and improving on its accuracy.
Ivan Garibay is an assistant professor in Industrial Engineering and Management Systems. He holds degrees that include a Ph.D. in computer science from UCF, and he is a director of UCF’s Artificial Intelligence and Big Data Initiative of CASL and a master’s program in data analytics.
“The presence of sarcasm in text is the main hindrance in the performance of sentiment analysis,” says Garibay. “Sarcasm isn’t always easy to identify in conversation, so you can imagine it’s pretty challenging for a computer program to do it and do it well. We 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.”
Garibay was joined by computer science doctoral student Ramya Akula and Brian Kettler, a program manager in DARPA’s Information Innovation Office (I2O).
Challenges of Text
“Sarcasm has been a major hurdle to increasing the accuracy of sentiment analysis, especially on social media, since sarcasm relies heavily on vocal tones, facial expressions and gestures that cannot be represented in text,” says Kettler. “Recognizing sarcasm in textual online communication is no easy task as none of these cues are readily available.”
Scientists from Garibay’s Complex Adaptive Systems Lab (CASL) rely on data science, network science, complexity science, cognitive science, machine learning, deep learning, social sciences, team cognition, and other approaches to address these challenges.
Akula is a graduate research assistant at CASL and a doctoral scholar. She holds a master’s degree in computer science from Technical University of Kaiserslautern in Germany and a bachelor’s degree in computer science from Jawaharlal Nehru Technological University in India.
“In face-to-face conversation, sarcasm can be identified effortlessly using facial expressions, gestures, and tone of the speaker,” Akula says. “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.”