基于相似度加强Louvain方法的复杂网络社区检测

Community detection based on similarity Optimized Louvain Method in complex networks

在针对复杂网络的社区检测问题中,基于模块度的社区划分方法Louvain Method(LM)迭代算法被广泛的应用,但是考虑到该算法在第一次迭代过程中的时间复杂度非常大,为了解决这个问题,文中引入了OLM方法,对复杂网络中的节点利用相似性的度量方法进行处理,从而优化整个网络结构,来更高效地使用该算法识别社区。实验结果表明,提出的OLM方法具有更高的效率和稳定性。

In the community detection problem for complex networks,the community division method,Louvain Method(LM)iterative algorithm based on modularity is widely used,but the time complexity of the algorithm in the first iteration process.Thus,in order to solve this problem,this paper introduces the OLM method to process the nodes in the complex network by using the similarity measurement method,so as to optimize the whole network structure and use the algorithm to identify the community more efficiently.Experiment results show that the proposed OLM method has higher efficiency and stability.