Author ORCID Identifier:
Date of Graduation
12-2025
Document Type
Thesis
Degree Name
Master of Science in Computer Science (MS)
Degree Level
Graduate
Department
Computer Science & Computer Engineering
Advisor/Mentor
Li, Qinghua
Committee Member
Jin, Kevin
Second Committee Member
Pan, Yanjun
Keywords
articficial intelligence; explainable AI (XAI); privacy; trustworthy
Abstract
Narrow Artificial Intelligence (AI) systems, defined for specific use cases, are used in all areas of life. In information-sensitive domains, these systems can engender trust through the inclusion of causality, fairness, privacy, and explainable decisions. AI systems that consider multiple values are troubled when definitions conflict with each other (e.g., model explanations may require the use of sensitive information, and privacy may desire the exclusion of such information). This thesis considers how to increase stakeholder trust in AI systems by (i) surveying underlying values influencing individual privacy behavior and (ii) mitigating disparate impact in multi-valued AI systems. This work offers initial insights into how values are prioritized by individuals, and how they can be reflected in AI systems. We explore a lesser researched area of multi-valued systems, privacy and model explainability.
Citation
Buckner, C. (2025). Exploring Trustworthiness in Privacy-Preserving Systems. Graduate Theses and Dissertations Retrieved from https://scholarworks.uark.edu/etd/6094
