Wayne State researchers developing trustworthy AI app to mitigate online-to-offline sexual violence

Sexual violence is a significant public health problem that affects millions of people in the U.S. Every 68 seconds, an American is sexually assaulted. Nearly seven out of every 10 victims are between the ages of 12-34.

Many abusive scenarios — sexual harassment, exploitation or abuse — are manifested online. As interpersonal relationships become increasingly facilitated through screens, it is important to understand how this form of communication shapes perceptions of appropriate behavior as well as the practice of giving and receiving consent to sexual activity. Researchers from Wayne State University and Oakland University are taking aim at this very subject.

Dongxiao Zhu, associate professor of computer science and the founding director of Wayne Artificial Intelligence (AI) Research Initiative at Wayne State, is the principal investigator on a project funded by a four-year, $196,892 National Science Foundation grant. Along with co-PI Rhongho Jang, assistant professor of computer science, and their Oakland University collaborators, this project will develop a framework to examine online dating and zero in on how users may be predisposed to becoming either perpetrators or victims of harm, even without their realization.

“In this AI for Social Good project, we will leverage trustworthy AI techniques in efforts to prevent online-to-offline sexual violence,” said Zhu. “We will develop a Siri alternative app — User Behavior Example (UBE) donator — to enhance trust in data donation from users based on textual and visual information to spot potential sexual predators.”

"While securing the data communication during data donation, we will preserve donators' privacy using state-of-the-art anonymization technologies,” said Jang.

The team’s goal is to gather and analyze a dataset of user behavior from multiple perspectives, including human-computer interaction, nursing and psychology, in order to develop preventative solutions.

Douglas Zytko, assistant professor in the Department of Computer Science and Engineering at Oakland University, is leading the effort for his institution, which received $403,107 of the total $600,000 award.

The grant number for this NSF award is 2211897.