基于核心节点的社区发现算法

Community discovery algorithm based on core nodes.Computer Engineering and Applications

针对基于全局的社区发现方法计算复杂度较高,基于局部的社区发现方法难以保证划分准确度的问题,本文提出了一种基于核心节点的社区发现算法。通过局部聚类系数和度计算网络中节点的优先级,以便更准确的选取核心节点,利用多层节点相似度判断其他节点与核心节点是否可以划分进同一小团体,可以提高节点划分准确性,最后,将紧密程度较大的小团体进行合并,得到划分结果。本文在真实的网络上进行验证,与GN算法和FN算法相比,本算法具有更好的划分准确性。

In order to solve the problem that the community discovery method based on the global community discovery is more complex and the local community discovery method is difficult to ensure the accuracy of the community discovery,a community discovery algorithm based on the core node is proposed in this paper.This paper calculates the priority of nodes in the network by local clustering coefficient and degree,so as to select the core nodes more accurately and determine whether other nodes and core nodes can be divided into the same small group by the similarity of multi-layer nodes,and the accuracy of the node partition can be improved.Finally,the small groups with larger tightness are merged,The result of division is obtained.This algorithm is verified in real network.Compared with GN algorithm and FN algorithm,this algorithm has better partition accuracy.