Community Structure Detection Algorithm Based on Rough Set
Zilu Cui, Wei Chu, Yuchen Fu · 2012
This paper proposes a new detection algorithm based on rough set. It uses information centrality as a measure of correlation between nodes. While dealing with the boundary nodes between communities, it uses upper and lower approximation subsets so as to better simulate the real world, then it clusters nodes to certain community and identifies the network to k communities. It identifies the ideal community structure according to modularity, and the value of k needs not to be given in advance. The algorithm is tested on two network datasets named Zachary Karate Club and College Football, and experimental result shows it has high accuracy rate.