Few mechanisms have been developed to clearly communicate data collection practices by websites to users at scale. We address this problem through automatically analyzing web traffic and creating usable privacy labels akin to nutrition labels. A proof-of-concept browser extension, Privacy Pioneer, is implemented to demonstrate the practical viability. At the browser-level, we can devise automated privacy analysis techniques that run locally while a user engages with the web. We then present analysis findings in a simple UI. In creating privacy labels for the web, we follow a shift towards privacy labels by the major operating systems iOS and Android. Packing a privacy analysis into a browser is a novel problem that requires working within its constraints and solving challenges of analysis accuracy, performance, and access to user data, among others.
"},{"label":"Degree","value":["Bachelor of Arts with Departmental Honors"]},{"label":"Author-supplied Keywords","value":["Data privacy","Software engineering","Machine learning"]},{"label":"Administrative Note","value":"Drupal Node ID: 24348