Researchers at the American Technion Society have developed the sarcasm Sentimental Interpretation GeNerator (Sarcasm SIGN), a system for interpreting sarcastic statements in social media. The team says the system is based on machine translation and turns sarcastic sentences into honest ones.
The researchers taught the system to produce accurate interpretations of sarcastic online statements by creating a database of 3,000 sarcastic tweets that were tagged with #sarcasm, in which each tweet was interpreted into a non-sarcastic expression by five human experts.
In addition, the team trained the system to identify words with strong sarcastic sentiments, and to replace them with strong words that reveal the true meaning of the text.
The system was examined by several human judges, who gave its interpretations high scores for fluency and accuracy.
The researchers say sentiment identification could be used in social, commercial, and other applications to improve communication between people and computers.
From American Technion Society
View Full Article
Abstracts Copyright © 2017 Information Inc., Bethesda, Maryland, USA