PhD Oral Exam - Didem Demirag, Information and Systems Engineering

When studying for a doctoral degree (PhD), candidates submit a thesis that provides a critical review of the current state of knowledge of the thesis subject as well as the student’s own contributions to the subject. The distinguishing criterion of doctoral graduate research is a significant and original contribution to knowledge.

Once accepted, the candidate presents the thesis orally. This oral exam is open to the public.

Abstract

Privacy is important both for individuals and corporations. While individuals want to keep their personally identifiable information confidential, corporations want to protect the privacy of their proprietary data in order not to lose their competitive advantage. Literature has extensively analyzed theoretical privacy and utilizing these primitives to address the need of privacy in real-world applications is useful for both individuals and corporations. We focus on different variations of a cryptographic primitive from the literature: secure multiparty computation (MPC) . MPC helps different parties compute a joint function on their private inputs, without disclosing them. In this dissertation, we look at real-world applications of MPC, and aim to protect the confidentiality of personal and/or proprietary data. Our main aim is to match theory to practical applications. The first work we present in this dissertation is a blockchain-based, generic MPC system that can be used in applications where personal and/or proprietary data is involved. Then we present a system that performs privacy-preserving link prediction between two graph databases using private set intersection cardinality (PSI-CA). The next use case we present again uses PSI-CA to perform contact tracing in order to track the spread of a virus in a population. The last use case is a genomic test realized by one time programs. Finally, this dissertation provides a comparison of the different MPC techniques and a detailed discussion about this comparison.