tnetwork.DCD.smoothed_graph

tnetwork.DCD.smoothed_graph(dynNetSN, alpha=0.9, match_function=<function jaccard>, threshold=0.3, **kwargs)

Smoothed graph approach

This approach is a naive implementation of the idea proposed in [1]. To sum up, at each snapshot, a new graph is create which is the combination of the graph at this step and a graph in which edges are present between any two nodes belonging to the same community in the previous step. Note than in the original paper, a method is proposed to greatly reduce the complexity of the solution, but this method is not implemented here.

Alpha is a parameter to tune how important is the weight of the current topology compared with previous partition.

The label attribution process is the same described in the paper XXX, see method simple_matching for details.

Internally, it calls the simple_matching method, the same parameters can be passed to it.

[1]Guo, C., Wang, J., & Zhang, Z. (2014). Evolutionary community structure discovery in dynamic weighted networks. Physica A: Statistical Mechanics and its Applications, 413, 565-576.

Parameters:

• dynNetSN –
• alpha – parameter setting relative importance of past VS current graph. 1: only current, 0: only previous

Returns: