Explore together how social media use and profiling shape recommendations and online content
Keywords: Recommender systems, profiling, social media
Somekone is a browser-based social media simulator that allows learners to explore and understand how social media platforms collect data, build user profiles, and recommend content. It visually resembles the feed of familiar social media image platforms, while making visible the processes that usually remain hidden.
In Somekone, every action—such as viewing, liking, commenting, or skipping an image—produces data.
The tool shows in real time how these small actions accumulate into digital traces, how they are used to construct user profiles, and how those profiles influence what content is recommended. Learners can examine similarities between profiles and connections between images. At the same time, it becomes clear that one user’s actions also affect the experiences of others and contribute to polarization.
Learners can experiment with the parameters of recommendation algorithms and observe how they shape the content feed, narrow or broaden perspectives, and create filter bubbles. In this way, Somekone supports understanding that social media is not a neutral information channel, but a data- and algorithm-driven system that steers attention and interaction.
The tool is designed for educational use:
it runs entirely in the browser, locally in the classroom, and does not collect or store personal data outside the class.
Key concepts: Data given, data traces, data inferred, profiling, engagement, recommendation, filter bubble, polarization, algorithmic influencing
Released in Spring 2024