Community Detection on Temporal Networks.

Temporal networks are widely used nowadays to represent dynamic systems in various contexts, from physics to biology, technology, economics and sociology. A well known example are social networks: the nodes represent the users, while the edges represent the connections between them, changing over time depending on their interactions. Community detection is an important analysis that can be done on the network in order to understand if the nodes are organized into groups or communities and how these evolve during time. This might be useful for real life application, for instance to discover disinformation campaigns on social networks. We analyse the current state-of-the-art algorithms. Specifically we focus on the trade-off between the stability of the algorithms over time and their ability to adapt to rapid changes in the communities structure.