一种采用邻居投票机制的重叠社区发现方法

发布时间：2018-04-10 08:05

本文选题：复杂网络　切入点：支持向量机　出处：《小型微型计算机系统》2014年10期

【摘要】：研究复杂网络中的社区结构,有助于发现网络结构和功能的关系,进而理解复杂网络的组成规律、预测复杂网络的行为.文章基于支持向量机的思想和LM(Louvain Method)非重叠社区发现算法,提出一种采用邻居投票机制的LM-NV(Louvain Method with Neighbor Voting)重叠社区发现方法,基本思想是保留非重叠社区的部分结构,采用一种基于局部信息的邻居投票机制仅对社区边界节点的社区隶属情况进行判别.LM-NV算法易于扩展到大规模复杂网络,同时不存在对社区个数的初始化问题.在基准测试网络和真实网络上的实验结果表明LM-NV算法不仅具有良好的时间效率,而且在社区发现准确度上优于其它代表性算法.
[Abstract]:The study of community structure in complex network is helpful to discover the relationship between network structure and function, to understand the composition of complex network, and to predict the behavior of complex network.Based on the idea of support vector machine (SVM) and the algorithm of LM(Louvain method non-overlapping community discovery, this paper proposes a LM-NV(Louvain Method with Neighbor mapping method based on neighbor voting mechanism. The basic idea is to preserve the partial structure of non-overlapping community.The neighborhood voting mechanism based on local information can only judge the community membership of the community boundary nodes. The algorithm is easy to be extended to large-scale complex networks and does not initialize the number of communities.The experimental results on benchmark network and real network show that the LM-NV algorithm not only has good time efficiency, but also is superior to other typical algorithms in community discovery accuracy.
【作者单位】：信息工程大学网络空间安全学院;数学工程与先进计算国家重点实验室;
【基金】：数学工程与先进计算国家重点实验室基金项目(2013A02)资助
【分类号】：TP393.09

【共引文献】