- Privately Estimating visualization using correlation
- Project Year:
2024
- REU Student (s):
Nigel Seymour | University of Maryland-Baltimore County MD
- Student 1 Institution:
University of Maryland-Baltimore County
- Project Mentor:
Anand Sarwate
- Project Mentor Area:
Electrical and Computer Engineering
- Project Abstract:
Researchers are interested in using data to demonstrate the results of health and human behavior of individuals. In sharing data, researchers have begun to shift to utilizing more data visualization techniques, and while these charts and graphs are visually more aesthetically pleasing, it is important to not violate the privacy of individuals who have reported their information. These data points must be protected using different techniques. While many focus on both qualitative support and quantitative results, the focus on the outputs must use algorithms to ensure that the data privacy remains secure and protected. I experimented using a linear correlation formula to understand how correlation works with different pieces of data from the US Census. Our goal for this project is to create differentially private correlation estimates.
