一种新的基于局部相似度的社区发现算法
发布时间：2018-10-31 13:14
【摘要】：社区发现是复杂网络领域的研究热点问题。为了提高复杂网络中划分社区结构的质量,提出了一种新的基于局部相似度的社区发现算法。首先,考虑到目前研究者们普遍基于共同邻居节点的自身特性来构建局部相似指标,通过引入节点对及其共同邻居间相互联络的亲密程度,定义了新的相似度指标;接着,基于网络节点相似度矩阵,结合改进的K-means算法对网络节点进行相似性聚类,实现网络的社区发现。在真实网络数据重构的网络上进行实验,结果表明,所提算法能够更准确、有效地发现复杂网络中的社区结构。
[Abstract]:Community discovery is a hot topic in the field of complex networks. In order to improve the quality of partitioning community structure in complex networks, a new community discovery algorithm based on local similarity is proposed. Firstly, considering that the researchers generally construct the local similarity index based on the characteristics of the common neighbor nodes, a new similarity index is defined by introducing the closeness degree between the nodes and their common neighbors. Then, based on the similarity matrix of network nodes, the improved K-means algorithm is used to cluster the network nodes to realize the community discovery. Experiments on real network data reconstruction show that the proposed algorithm is more accurate and effective in finding community structures in complex networks.
【作者单位】： 南京邮电大学自动化学院;
【基金】：教育部人文社会科学研究规划基金(15YJAZH016) 江苏省普通高校研究生创新计划(SJZZ16_0151)资助项目
【分类号】：O157.5
[Abstract]:Community discovery is a hot topic in the field of complex networks. In order to improve the quality of partitioning community structure in complex networks, a new community discovery algorithm based on local similarity is proposed. Firstly, considering that the researchers generally construct the local similarity index based on the characteristics of the common neighbor nodes, a new similarity index is defined by introducing the closeness degree between the nodes and their common neighbors. Then, based on the similarity matrix of network nodes, the improved K-means algorithm is used to cluster the network nodes to realize the community discovery. Experiments on real network data reconstruction show that the proposed algorithm is more accurate and effective in finding community structures in complex networks.
【作者单位】： 南京邮电大学自动化学院;
【基金】：教育部人文社会科学研究规划基金(15YJAZH016) 江苏省普通高校研究生创新计划(SJZZ16_0151)资助项目
【分类号】：O157.5
【相似文献】
相关期刊论文 前8条
1 冯成进;;0—1规划新算法的改进[J];曲阜师范大学学报(自然科学版);1988年02期
2 欧阳普仁,杨叔子;一种改进的Marple算法[J];南京理工大学学报(自然科学版);1988年01期
3 刘新亮;张涛;郭波;;基于分布估计算法的备件优化配置[J];系统工程理论与实践;2009年02期
4 杨大志;崔勇;闫忠文;;博弈的期望算法[J];河北科技师范学院学报;2005年04期
5 陈煜聪;杨斌;杜海峰;邵，
本文编号：2302325
本文链接：https://www.wllwen.com/kejilunwen/yysx/2302325.html
