Welcome to the web self of the project Algorithms Exposed (AKA ALEX).
ALEX is meant to provide researchers, advocates, policymakers and journalists with reliable and accessible algorithmic auditing methods and data. In our view, it empowers users to independently monitor, compare, and reflect upon their information diets. It is innovative because:
- It puts the user in the driver seat, by giving individuals full control of data extraction patterns. In other words, users can decide at any time if and what data to volunteer and to what research project. They can also decide to withdraw their participation whenever they are so inclined.
- It supports data protection and user privacy: it is fully GDPR-compliant, it does not interfere with the user privacy settings and observes only publicly available content appearing in the Facebook news feed, not individual profiles or pages.
- Users retain data sovereignty. Users decide what selected portions of their social media experience they want to share; they own their data and can at any time see how data are used in aggregate analyses.
- It is open-source! You can download and customise it. Although we use Facebook as a test case, our methodology can be applied to any other social media platform.
- By enabling self-awareness, it promotes a healthy information diet. By unpacking the functioning of personalisation algorithms, it fosters a responsible use of social media.