期刊
JOURNAL OF BIONIC ENGINEERING
卷 20, 期 3, 页码 1175-1197出版社
SPRINGER SINGAPORE PTE LTD
DOI: 10.1007/s42235-022-00303-z
关键词
Bionic algorithm; Complex network; Community detection; Harris hawk optimization algorithm; Opposition-based learning; Levy flight; Chaotic maps
The purpose of community detection in complex networks is to identify the structural location of nodes. This paper proposes different Harris Hawk Optimization (HHO) algorithm methods that aim to balance exploitation and exploration for community detection in social networks. The finding shows that the IHHOOBL method has better detection accuracy than IHHOLF and IHHOCM.
The purpose of community detection in complex networks is to identify the structural location of nodes. Complex network methods are usually graphical, with graph nodes representing objects and edges representing connections between things. Communities are node clusters with many internal links but minimal intergroup connections. Although community detection has attracted much attention in social media research, most face functional weaknesses because the structure of society is unclear or the characteristics of nodes in society are not the same. Also, many existing algorithms have complex and costly calculations. This paper proposes different Harris Hawk Optimization (HHO) algorithm methods (such as Improved HHO Opposition-Based Learning(OBL) (IHHOOBL), Improved HHO Levy Flight (IHHOLF), and Improved HHO Chaotic Map (IHHOCM)) were designed to balance exploitation and exploration in this algorithm for community detection in the social network. The proposed methods are evaluated on 12 different datasets based on NMI and modularity criteria. The findings reveal that the IHHOOBL method has better detection accuracy than IHHOLF and IHHOCM. Also, to offer the efficiency of the , state-of-the-art algorithms have been used as comparisons. The improvement percentage of IHHOOBL compared to the state-of-the-art algorithm is about 7.18%.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据