Skip to main content
Fig. 1 | Applied Network Science

Fig. 1

From: Identifying and exploiting homogeneous communities in labeled networks

Fig. 1

a Number of communities and b size of the largest community as functions of α. c-e Modularity-Purity as a function of α, where the lines represent the average values over ten executions, and the surrounding areas are the interquartile ranges. EVA works well until the decrease of modularity values (i.e., the red trends): for instance, in G+, it starts from α=0.6, coherent with a relevant increase of partition cardinality

Back to article page