Zeyu Ding
Binghamton University, NY·Computer Science
Companies and government agencies hold databases full of sensitive personal information, such as tax filings, travel records, purchase histories and salaries. The same data could improve operations, guide policy and describe society.
- Fields
Computer Science
Mathematics & Statistics
- Topics
differential privacy
formal privacy
data privacy
- Grants
- 1 known grant, through Apr 2028
What they work on
- Build automated tools that analyze an SQL database schema and help users interactively define a customizable privacy model specifying which data elements need differential-privacy protection, and which variant.
- Develop automated reasoning about SQL queries that determines how much noise must be injected and produces privacy-preserving query execution plans matched to the chosen privacy model.
- Optimize accuracy of privacy-preserving query answers and deliver an open-source, machine-checked privacy middleware compatible with existing SQL databases.
More about the research
Differential privacy is a mathematical approach that adds carefully calibrated random noise to analysis results, so that no individual's information can be confidently revealed while the answers stay useful. Today it is hard to apply without deep technical expertise. Earlier systems also used rigid, pre-set privacy models that limited what they could support.
Why it matters
An easy-to-use privacy layer for SQL databases could let government agencies and companies share useful statistics from tax, travel, customer or employee records without exposing individuals. It could make formal privacy protection available to organizations that lack specialized experts, and could support policy analysis built on sensitive data.
Known grants
From NIH and NSF research grants. Labs often have other funding we can't see.
Closest labs
Christopher Bryan
Arizona State University Campus Immersion, AZ·Computer Science
Develop a visualization grammar, a way to specify chart designs without low-level programming, that supports privacy constraints and data sanitization actions.
Andrew Pavlo
Carnegie Mellon University, PA·Computer Science
Develop automatic code transformations that speed up user-defined functions (UDFs) inside SQL queries without requiring changes to application code.
Indrakshi Ray
Colorado State University-Fort Collins, CO·Computer Science
Develop natural language processing models for cross-genre entailment that connect the meaning of privacy laws to technical software specifications, in order to identify privacy vulnerabilities in health data handling.
The school
Binghamton University: 14 labs on PI Pages, $161M a year on research, 132 research doctorates a year.
Reach out
A start for your email. Fill in the brackets.