A tremendous amount of data is generated every second, every day by billions of users powering social media. Recent statistics indicate that every minute Instagram users post nearly 220,000 new photos, YouTube users upload 72 hours of video, and Facebook users share nearly 2.5 million pieces of content. Organizing and making sense of big data has become one of today’s major challenges in machine learning and data mining. Karbasi’s research, recognized by Google, aims at extracting a small subset of representative elements in order to obtain a faithful description of the whole massive data.