Have you ever wondered how Facebook collects all the data it has to feed you with the content it presumes you’ll like and keep you coming back for more? Well, now there’s an app that can answer these questions.
Available for free, Data Selfie is an open-source Chrome extension that helps you discover how machine learning algorithms track and process your Facebook activity, and gain insights about your personality and habits.
To accomplish this, the nifty extension monitors your Facebook interactions for patterns and then crunches the collected data into insightful reports.
Data Selfie essentially tracks your activity – what you look at, how long you look at it, what you like, what you click and what you type – and then applies natural language processing and machine learning algorithms from IBM Watson and the University of Cambridge to turn this data into insight.
The extension comes with a handy dashboard that shows your aggregated Facebook activity in a timeline, conveniently broken down with color coding to highlight different aspects about your data usage.
In addition to this, the Data Selfie dashboard also includes insight into what posts you’ve spent most time on – both for friends and liked pages. In a creepily fascinating way, the extension also uses predictive analytics to guess stuff like your political affiliations as well as shopping and nutrition preferences.
To prevent ill-intended individuals from obtaining the information it collects about you, Data Selfie keeps your data locally – only on your own machine – and never stores anything on external servers.
As part of its initiative to promote internet transparency, creator Data X has made the code for Data Selfie available on GitHub for curious developers to peruse. Head to this repository for more details.
Start learning how Facebook’s algorithms collect and interpret your activity patterns, and get Data Selfie from the Chrome Web Store here.
Pssst, hey you!
Do you want to get the sassiest daily tech newsletter every day, in your inbox, for FREE? Of course you do: sign up for Big Spam here.