JNS
The program, funded by the U.S. military, can determined whether short texts are "positive or negative."
A scientist at Ben-Gurion University of the Negev has designed an AI system capable of identifying "social norm violations" in text samples.
With U.S. military funding, professor Yair Neuman and engineer Yochai Cohen built the system using GPT-3, zero-shot text classification and automatic rule discovery.
They trained the system to identify 10 social emotions: competence, politeness, trust, discipline, caring, agreeableness, success, conformity, decency and loyalty. The system successfully classified texts into one of these 10 groups, and defined them as positive or negative.
The system was tested on two massive datasets of short texts and empirically proved the validity of the models, according to a statement from BGU.
The U.S. Defense Department's Defense Advanced Research Projects Agency (DARPA) commissioned The Computational Cultural Understand (CCU) program to create cross-cultural language understanding technologies to improve situational awareness and interactional effectiveness. Cross-cultural miscommunication not only derails negotiations, but also can be a contributing factor leading to war, according to DARPA's explanation of the program.
The findings were published recently in the journal Scientific Reports.
"This is a preliminary work, but it provides strong evidence that our approach is correct and can be scaled up to include more social norms," said Neuman, who heads The Functor Lab in the Department of Cognitive and Brain Sciences at BGU.
