Skip to content

Studying existing research developments in ε-differentially private algorithms on structured and unstructured data using Google’s Differential Privacy Library (C++)

Notifications You must be signed in to change notification settings

guptajay/differential-privacy-website

Repository files navigation

Differential Privacy in Machine Learning

This was a research project under NTU’s URECA (Undergraduate Research Experience on Campus) programme. I was presented URECA Undergraduate Research Excellence Award in recognition of outstanding research accomplishments with Distinction in URECA Undergraduate Research Programme AY2019-20.

Differential privacy makes it possible for tech companies to collect and share aggregate information about user habits, while maintaining the privacy of individual users.

In this project, we studied existing research developments in ε-differentially private algorithms on structured data using Google’s Differential Privacy Library and refine it so it can be used in real world production data.

Screenshot

This was a year long project, with a lot of unique experiences. In the beginning, we started with reading reseeach papers on Differential Privacy. In the development of the website shown above, I learnt how to operate Ubuntu Servers, use PostgreSQL and build websites with Flask.

About

Studying existing research developments in ε-differentially private algorithms on structured and unstructured data using Google’s Differential Privacy Library (C++)

Topics

Resources

Stars

Watchers

Forks