Local-structure-based community detection in firm network
Liang Zhang1, YulingLiao2, Xing Cao1,GuodongHao3
COMPUTER MODELLING & NEW TECHNOLOGIES 2014 18(12B) 253-257
1Business School, Central South University, Changsha 410083, China
2Department of Legal and Public Administration, Hunan University of Finance and Economics, Changsha 410205, China
3School of Information Science and Engineering, Central South University, Changsha 410083, China
The community detection algorithm based on label propagation could discover latent community structure in a complex network by propagating the label of node between neighboring nodes. Due to the uncertainty and randomness involved in the propagation process, the output community structure is often unstable and lacks precision. The Local-Structure-Based Community detection algorithm propagates the label as an entity instead of propagates the labels of individual nodes, while defying the influence of label on the basis of the local structure in the network. Experiments shows that, applied on complex network data-bases, the algorithm can output high quality community structure, and the output was stable.