An Improved SGN Algorithm Research for Detecting Community Structure in Complex Network
In order to make more accurate partition community structure of complex networks, this paper puts forward a new community partition algorithm. The basic idea of the algorithm depends on node similarity, and it deletes the link whose similarity is the smallest every time, then takes modularity Q as the judging standard. Computing the corresponding modularity when network occurs into pieces, and the module structure is the ultimate community structure when Q reaches its peak. This algorithm not only improves the accuracy of the original algorithms, but also makes sure that the community structure has a better quantification. When the new algorithm is applied to the complex networks, we finally find that the algorithm is effective and feasible.
Keywordscomplex network community structure node similarity modularity
Unable to display preview. Download preview PDF.
- 6.Newman, M.E.J., Girvan, M.: Finding and evaluating community structure in networks. Phys. Rev. E 69, 026113 (2004)Google Scholar
- 8.Ting, L., Baoqing, H.: Based on clustering analysis in complex network community detection. Journal of Complex Systems and Complexity Science 4(l), 29–35 (2007) (in Chinese)Google Scholar
- 9.Yawen, J., Caiyan, J., Jian, Y.: The network community detection algorithm based on node similarity research. Journal of Computer Science 38(7) (2011) (in Chinese)Google Scholar
- 10.Newman, M.E.J.: Phys. Rev. E 69, 066133 (2004)Google Scholar
- 11.Xutao, W., Guanrong, C., Hongtao, L.: A very fast algorithm for detecting community structures in complex networks. Physica A384, 667–674 (2007) (in Chinese)Google Scholar
- 12.Newman, M.E.J.: Scientific collaboration networks. Phys. Rev. E 64, 016132 (2001)Google Scholar