Authors: Amir Reza Asadi, Joel Appiah, Taiwo Peter Akinemi, Hazem Said (University of Cincinnati) This project is about creating a testbed for idnetifying naunced aspects of user beahvior during system design and development. Extended Abstract at USENIX SOUPS 2025 https://www.usenix.org/conference/soups2025/presentation/asadi-poster
• Creates a behavioral testbed with three persona categories: regular users, threat actors, and security practitioners • Introduces three evaluation metrics: Persona Fidelity Index, Behavioral Diversity Index, and Vulnerability Detection Rate • Aims to proactively discover security vulnerabilities and privacy risks through automated persona-based testing
A fundamental challenge in system design and security research is the inability to test across the full spectrum of user diversity. While planning phases identify the need to consider various user groups—including those with different technical expertise, cultural backgrounds, accessibility requirements, or usage contexts—researchers often cannot directly access or engage these diverse populations during testing. This gap between recognized user diversity needs and actual testing capabilities creates blind spots in security and usability assessments.
The testbed addresses this critical limitation by simulating diverse user behaviors, enabling researchers to proactively identify vulnerabilities and privacy risks across user groups that would otherwise remain untested. Additionally, the research aims to validate the accuracy and effectiveness of persona-based modeling in representing real user behaviors during security assessments.
Research Overview
npm run dev
Open http://localhost:3000 with your browser to see the result.
You can start editing the page by modifying app/page.tsx. The page auto-updates as you edit the file.
This project uses next/font to automatically optimize and load Geist, a new font family for Vercel.