Title:FinTech, Data Analysis, and Privacy Preservation
Presenter:Ning Cai, Department of Industrial Engineering and Decision Analytics, Hong Kong University of Science and Technology
Invited by: Li Ping
Time:2019.4.24 10:00-12:00
Location:A622
Abstract
Many data are sensitive in areas such as finance, economics, political science, and life science. How to protect individual privacy when collecting and analyzing data has become even more significant and has raised growing public concerns in the age of big data because a huge amount of personal data are being generated and used almost freely every day. We propose an ER (encryption and recovery) algorithm that allows a central administration to do statistical inference based on the encrypted data, while still preserving each party's privacy even for a colluding majority in the presence of cyber attack. Theoretically, we essentially establish a general framework for privacy-preserving statistical inference, which can be viewed as the sensitive data based counterpart of traditional statistical inference assuming availability of the data. We demonstrate the applications of our algorithm to linear regression, logistic regression, maximum likelihood estimation, estimation of empirical distributions, and estimation of quantiles. Moreover, our algorithm can help to address another practically significant issue -- privacy preservation for distributed statistical inference when data are allocated to different parties who are unwilling to share their own data with others. Our algorithm is a promising “technology” that can be applied to overcome the difficulties of data analysis with privacy preservation not only in financial industry but also in other areas concerning people's privacy rights in the era of big data.
