摘要
针对现有的社团检测算法存在准确度低、没有充分考虑到有向网络的方向特性等问题,提出一种改进的能够适用于有向网络的CNM(Newman贪婪算法)社团检测算法。在算法设计中引入基于拓扑结构信息的有向网络节点相似度算法,并重新定义模块度增量函数ΔQs。使用一个计算机生成网络和两个实际网络对算法进行了测试并与已有算法进行比较。实验结果表明,文章提出的算法能够有效地检测出有向网络中的社团结构。
Through the study of the existing community detection algorithm, problems of low accuracy rate, ignore the direction of the edge are found, and an improved CNM algorithm based on similarity in directed networks is presented. The new algorithm introduces a similarity algorithm based on topological information to calculate similarity between the pairs of nodes in the given direct networks and defines a new AQ, function for CNM algorithm. The performance of the proposed algorithm is tested and compared with other algorithms on one computer-generated network and two real networks. Experimental results show that the algorithm presented in this paper is rather efficient to detect communities of directed networks.
出处
《微型机与应用》
2017年第3期19-22,共4页
Microcomputer & Its Applications
关键词
社团检测
有向网络
CNM算法
节点相似度
community detection
directed networks
CNM algorithm
node similarity
