Community Detection with Self-Adapting Switching Based on Affinity

Figure 5

The detection results for various real-world networks. For the karate network (a), the result is identical with the original. But for the dolphin network (b), the left community in the original has been divided into two communities, which are marked in red and green, respectively. In the football team network (c) and the political books network (d), the detected communities are illustrated by dashed lines and the original partitions are distinguished by different colors. The parameters , , and .

(a)

(b)

(c)

(d)