Digital Social Dynamics

Analysis of Social Embedding

Humans are social creatures and connectedness to others is relevant for our well-being. To understand the link between well-being and social interactions, it is important to objectively measure these interactions – a challenge that we want to solve with our data donation method.
Dona is a platform, on which the participant donates their anonymized social media interaction data from a social platform such as Facebook, Instagram or Whatsapp. We extract only meta-data from these files (timestamps and word counts) so that the data are completely de-identified.

Our goal is to understand how social interaction patterns are associated with human well-being.
We use time series analysis to study changes of social interactions over time, e.g. how do the times or the intensity of social interactions change. 
In addition, by using machine learning methods we aim to identify interaction strategies and patterns that are associated with different psychological characteristics.

To reward the study participants for making these analyses possible, we provide personalized feedback on their messaging behavior in the form of intuitive visualizations. The participants can learn whom they writes the most, when are they most active or how fast do they reply.

This research was funded by the "Empathische Künstliche Intelligenz" (EKI) grant, FKZ 01IS20046 from 01.12.2020-30.11.2024.