Most research focuses on generalization or the assumption of a single linguistic framework. Rios, however, will apply grant funding to develop NLP models specific to geographic populations, ethnic communities, and other population-specific groups.
“NLP is about developing tools to understand language, but we don’t have language without people,” explained Rios, who is also associate director of UTSA’s Cyber Center for Security and Analytics. . “My specific goal is to develop NLP tools for biomedical and social applications, which I believe can have real impact and hopefully improve people’s lives.”
The Rios project has three goals: to determine methods to detect correlations between community language and NLP model performance; develop a tool to help community leaders find the potential harms of specific NLP models and help them decide if those harms are relevant to that community; and to find processes and opportunities to improve the performance of the NLP model.
“I’m interested in understanding how these patterns can impact people and what biases exist within them,” Rios said. “There’s a lot of hype around NLP and artificial intelligence, but I think it’s important to avoid the hype and understand the potential damage caused by NLP models when deployed. “
Rios comes to this project with a background and multidisciplinary research interests in NLP, machine learning, and computational social sciences. He received his introduction to NLP during his doctoral program in computer science at the University of Kentucky, where his academic advisor was Ramakanth Kavuluru, experienced researcher in NLP. The many applications of NLP and its impact on people attracted Rios to the field.