Science

Researchers are developing artificial intelligence that can detect irony on social media



ANI |
Has been updated:
May 11, 2021 23:07 IST

Washington [US]May 11 (ANI): Appropriately understand and respond to customer feedback Social media The platform is very important to the brand and may have been a bit easier thanks to new computer research. Science Researchers at the University of Central Florida Irony Detector.
Social media has become a major means of communication for individuals and companies looking to market and sell their products and services. Properly understand and respond to Twitter, Facebook and other customer feedback Social media The platform is important for success, but it’s incredibly labor-intensive.
That’s where sentiment analysis comes in. The term refers to an automated process that identifies emotions (positive, negative, neutral) associated with the text. Artificial intelligence refers to logical data analysis and response, but sentiment analysis is similar to correctly identifying emotional communication. The UCF team has developed a technology for accurate detection Irony so Social media Sentence.
The team’s findings were recently published in the journal Entropy.
Effectively, the team taught computer models to find patterns that often show. Irony And in combination with teaching it to the program, it correctly selects the cue words in the sequence that is likely to be directed. Irony.. They taught the model to do this by feeding a large dataset to the model, and then checked its accuracy.
“The presence of Irony It is a major obstacle to the performance of sentiment analysis in the text, “said Ivan Garibay ’00 MS’04 PhD, an assistant professor of engineering.
“Irony isn’t always easy to identify in a conversation, so you can imagine that it’s pretty hard for a computer program to do it and do it well. We have multi-headed self-attention and gated regression. Using the unit, we have developed an interpretable deep learning model. The multi-head self-attention module helps identify important ironic cue words from the input, and the iterative unit better classifies the input text. To learn the long-range dependencies between these cue words. “

Team including computer Science A PhD student, Ramya Akula began addressing this issue under a DARPA grant that supports computational simulation of an organization’s online social behavior programs.
“Irony is a major obstacle, especially for improving the accuracy of sentiment analysis. Social mediaSince Irony Brian Kettler, Program Manager at DARPA’s Information Innovation Office (I2O), said: Irony Online text communication is not an easy task as these clues are not readily available. “
This is one of the challenges of Garibay’s Complex Adaptive Systems Lab (CASL). studying. CASL is study Complex phenomena such as world economy, world information environment, innovation ecosystem, sustainability, social and cultural dynamics and evolution
CASL scientist study These issues with the data ScienceCommunication network Science,complexity ScienceCognition ScienceMachine learning, deep learning, social Sciences, team recognition, among other approaches.
“In face-to-face conversations Irony It’s easy to identify using facial expressions, gestures, and speaker tone, “says Akula. Irony None of these clues are readily available, so text communication is not an easy task. Especially with the rapid increase in Internet usage Irony Detection via online communication from social networking platforms is much more difficult. “
Garibay is an assistant professor of industrial engineering and management systems.He has several degrees, including a PhD in Computers Science From UCF.Gullivey is the director of UCF Artificial intelligence Big Data Initiative for Master’s Program in CASL and Data Analysis.
His areas of research include complex systems, agent-based models, information and false alarm dynamics. Social media, Artificial intelligence and machine learning. He has over 75 peer-reviewed papers and over US $ 9.5 million in funding from various state institutions. (ANI)



https://www.aninews.in/news/researchers-develop-artificial-intelligence-that-can-detect-sarcasm-in-social-media20210511230743/ Researchers are developing artificial intelligence that can detect irony on social media

Back to top button