基于结构近似度的社交网络聚类  

Social network clustering analysis based on structural approximation

作　　者：王韫烨 孔珊 李亚伦[2] Wang Yunye;Kong Shan;Li Yalun(College of Information Science & Technology,Zhengzhou Normal University,Zhengzhou 450044,China;School of Electronic and Information Engineering,Tianjin Polytechnic University,Tianjin 300387,China)

机构地区：[1]郑州师范学院信息科学与技术学院,河南郑州450044 [2]天津工业大学电子与信息工程学院,天津300387

出　　处：《南京理工大学学报》2020年第2期230-235,共6页Journal of Nanjing University of Science and Technology

基　　金：国家自然科学基金(61572447,61972456);河南省科技攻关项目(162102310238)。

摘　　要：针对基于结构近似度的聚类算法无法解决非对称网络聚类的问题,该文根据社交网络的特点,提出了基于结构近似度的有向社交网络聚类算法,通过将社交网络抽象为图结构,将网络聚类问题看成图论中的子图划分问题,实现了对社交网络的准确聚类分簇,且分簇复杂度较低。使用C++语言编程实现该算法,通过自定义有向网络数据集和标准数据集的测试表明,该算法对社交网络结构的划分较为准确,且能鉴别离群节点和枢纽节点。In view of that the clustering algorithm based on structural approximation can not solve the clustering problem of the asymmetric network,a directed clustering algorithm based on structural approximation is proposed here.The social network is studied as a graph structure,and the network clustering problem is regarded as a sub-graph division to realize the clustering of directed graphs with low complexity.The algorithm was achieved by C++programming,and the customized directed network datasets and standard datasets are used to test the proposed algorithm.The experimental results show that the algorithm for the network structure is more accurate and can identify the outliers and hub points.

关 键 词：社交网络 有向图 网络聚类 结构近似度 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]