Partitioning weighted social networks based on the link strength of nodes and communities

被引：0

作者：

Liu, Miao-Miao ^{[1
]}

Guo, Jing-Feng ^{[2
]}

Chen, Jing ^{[2
]}

机构：

[1] Northeast Petroleum University, Daqing,Heilongjiang,163318, China

[2] College of Information Science and Engineering, Yanshan University, Qinhuangdao,Hebei,066004, China

来源：

Journal of Information Hiding and Multimedia Signal Processing | 2018年 / 9卷 / 01期

基金：

中国国家自然科学基金;

关键词：

D O I：

暂无

中图分类号：

TN911 [通信理论];

学科分类号：

081002 ;

摘要：

The algorithm CD WLS is proposed for partitioning weighted social networks reasonably and effectively. Firstly, the link strength between nodes based on their common neighbors is defined as their weighted similarity. Then nodes are clustered fast and the initial partition of the network is achieved. Finally, closely connected communities would be merged on the basis of the definition of the link strength between communities so as to optimize the initial partition and get a more accurate division result. Experiments are carried out on many artificial and real weighted networks using the weighted modularity as the evaluation criterion to verify the effectiveness and correctness of the algorithm proposed. Results show that the similarity index of weighted link strength defined in the paper is superior to WCN, WAA and WRA. Meanwhile, the speed and accuracy of CD WLS algorithm are improved greatly compared with the WGN algorithm. Furthermore, the algorithm proposed can achieve higher accuracy for community partition in weighted networks than Lu Algorithm and CRMA algorithm. © 2018, Ubiquitous International. All rights reserved.

引用

页码：21 / 32