Publication Date

Spring 2019

Degree Type

Master's Project

Degree Name

Master of Science (MS)


Computer Science

First Advisor

Thomas Austin

Second Advisor

Ben Reed

Third Advisor

Fabio Di Troia


Data confidentiality, Declassification, Downgrading poli- cies, Dimensions of declassification


This research addresses the issues with protecting sensitive information at the language level using information flow control mechanisms (IFC). Most of the IFC mechanisms face the challenge of releasing sensitive information in a restricted or limited manner. This research uses faceted values, an IFC mechanism that has shown promising flexibility for downgrading the confidential information in a secure manner, also called declassification.

In this project, we introduce the concept of first-class labels to simplify the declassification of faceted values. To validate the utility of our approach we show how the combination of faceted values and first-class labels can build various declassification mechanisms.