The detection of communities within networks or graphs has clear applications in domains ranging from sociology to epidemiology. While static networks represent interactions as the relations between nodes and edges, temporal networks show how the interactions between nodes change dynamically over time (by replacing edges with events). Most approaches to community detection in temporal networks transform the network into a sequence of static networks defined on a discrete time grid. However, these approaches may not accurately capture the impact of the temporal evolution of networks on network behavior and topography. In contrast, Bovet et al. propose an approach wherein they cluster the flow of random walkers co-evolving on the network and limited by the activation time of the edges. They capture the temporal evolution of networks with a forward partition and backward partition.