This course aims to teach learners how to use an automated framework for privacy policy analysis called Polisis, which enables scalable, dynamic, and multi-dimensional queries on natural language privacy policies. The course covers the development of a privacy-centric language model and a hierarchy of neural-network classifiers to analyze privacy practices in privacy policies. Students will learn how to use Polisis for structured querying, such as assigning privacy icons, and free-form questioning through an application called PriBot. The teaching method includes demonstrations of the applications and their utility. This course is intended for companies, users, researchers, and regulators who are interested in understanding and analyzing privacy policies efficiently and effectively.
Overview
Syllabus
USENIX Security '18 - Polisis: Automated Analysis and Presentation of Privacy Policies...
Taught by
USENIX