用于社区发现的LPA_LRDC标签传播算法

LPA_LRDC Tag Propagation Community Discovery Algorithm Based on Optimized LeaderRank

由于传统的LPA算法,在节点标签更新的顺序以及标签传播过程中存在较大的随机性,给社区发现的准确性和稳定性造成了很大的影响.本文提出LRDC(Leader Rank algorithm considered degree and clustering coefficient)算法并用其来衡量节点的重要性,然后按照节点的重要性大小排序作为LPA算法中初始化节点标签的依据,并在标签传播过程中综合考虑节点重要性以及邻居标签的数量提出LPA_LRDC(Label Propagation Algorithm based on LRDC)标签传播社区发现算法.通过在人工和真实的网络数据集上的实验结果表明,本文提出的标签传播社区发现算法能够显著的提高社区发现的准确性和稳定性.

Because of the large random in the order of node label updating and the label propagation process in the traditional LPA algorithm, the accuracy and stability of the community discovery is influenced largely. In this paper, the LRDC （LeaderRank algorithm considered degree and clustering coefficient ） algorithm is proposed and used to measure the importance of nodes, and the nodes sorted according to the importance of node are the basis for the initialization of node labels in the LPA algorithm, and the importance of nodes and the number of neighbor tags are considered in the process of tag communication, and then the LPA_LRDC （ Label Propagation Algorithm based on LRDC） tag communication community discovery algorithm is proposed. Experimental results on both artificial and real network data sets show that the proposed algorithm can significantly improve the accuracy and stability of community detection.