A Link Strength Based Label Propagation Algorithm For Community Detection

被引：6

作者：

Lakhdari, Abdallah ^{[1
]}

Chorana, Aicha ^{[1
]}

Cherroun, Hadda ^{[1
]}

Rezgui, Abdelmounaam ^{[2
]}

机构：

[1] Univ Amar Telidji, LIM, Laghouat, Algeria

[2] New Mexico Inst Min & Technol, Cloud Comp & Big Data Lab C2BD, Socorro, NM USA

来源：

PROCEEDINGS OF 2016 IEEE INTERNATIONAL CONFERENCES ON BIG DATA AND CLOUD COMPUTING (BDCLOUD 2016) SOCIAL COMPUTING AND NETWORKING (SOCIALCOM 2016) SUSTAINABLE COMPUTING AND COMMUNICATIONS (SUSTAINCOM 2016) (BDCLOUD-SOCIALCOM-SUSTAINCOM 2016) | 2016年

关键词：

Social networks; Community detection; LPA; K-neighborhood; COMPLEX NETWORKS; PREDICTION;

D O I：

10.1109/BDCloud-SocialCom-SustainCom.2016.61

中图分类号：

TP3 [计算技术、计算机技术];

学科分类号：

0812 ;

摘要：

The Label Propagation Algoritlun (LPA) is a fast algorithm for community detection. This algoritlun has proved its efficiency and scalability even in large social networks. It is based only on updating the label of each node by the most frequent label within its neighbors. However, in case of many maximal labels, a random choice is made. This reduces the performance of the algorithm. In this paper, we propose a new version of LPA, called Link Strength-based LPA (LS-LPA), that addresses this issue. In our approach, we rank labels according to the link strength of each neighbor. The strength of links is quantified via k-neighborhood which includes common neighbors and multi-step neighbors. Experiments on different networks with different sizes show that our solution improves LPA's performance significantly.

引用

页码：362 / 369

页数：8

共 25 条

[1] Statistical mechanics of complex networks [J].