CAREER: Modeling and Mitigating Privacy Risks to Student Data at Higher Education Institutions

Digital technologies are ubiquitous at institutions of higher education. Their use generates enormous amounts of data that could improve teaching and learning processes. However, the accidental leakage or intentional misuse of such data can create severe privacy and safety risks for students and others at these institutions. Thus, the goal of this project is to develop a systematic method for data governance that incorporates formal risk assessments and proactive mitigation techniques. The project seeks to leverage institutional data to advance science without harming data subjects. This project aims to improve student privacy protections and education research pertaining to data security. The project focuses on three activities. First, it formalizes the privacy risk modeling process at institutions of higher education and creates a threat ontology. Second, it develops two classes of novel mitigation techniques to privacy attacks: one based on censoring internal representations of data by machine learning models, and another based on causality-aware, differentially private synthetic data generation. Third, following a participatory design approach, it develops a web-based, visual, and interactive privacy risk assessment tool that builds on the threat ontology and mitigation techniques. In addition, the project team is testing the usability of the threat modeling tool and deploying it at three universities along with conducting training sessions with data stewards and privacy officers. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria. NSF Award ID: 2544663 | Program: 01002627DB NSF RESEARCH & RELATED ACTIVIT,01002930DB NSF RESEARCH & RELATED ACTIVIT,01003031DB NSF RESEARCH & RELATED ACTIVIT | Principal Investigator: Rakibul Hasan | Institution: Arizona State University, SCOTTSDALE, AZ | Award Amount: $391,490 View on NSF Award Search: https://www.nsf.gov/awardsearch/show-award/?AWD_ID=2544663 View on Research.gov: https://www.research.gov/awardapi-service/v1/awards/2544663.html