Fuzzy duocentric community detection model in social networks

Abstract

a b s t r a c tThe main goal of this paper is to present a clustering model to identify duocentric communities in the complex networks. A duocentric community is built around two central nodes which are as close as possible to other nodes, while the central nodes are connected enough to each other to shape the center of the community. To detect such communities, we develop a new objective function based clustering model. The network's nodes are assigned to the duocentric communities by the type-2 fuzzy numbers which indicate the degrees of belonging to the communities by upper and lower membership values. Generated interval type-2 fuzzy membership values by our proposed model are able to determine how much each node belongs to the both central nodes and how it is shared among communities. Also, the compatible verification index with the proposed model is introduced to evaluate and compare the results of the proposed model with the existing approach in the literature. Finally, the performance of the proposed algorithm is validated by detecting duocentric communities in three artificial networks and two real social networks.

A possibilistic framework for the detection of terrorism‐related Twitter communities in social media

“…Some fuzzy community detection algorithms calculate the possibility of each node belonging to every community, such as in other works. [23][24][25][26] The authors of these approaches provide a fresh idea for the understanding of community detection. Therefore, coupling relationship and content information in social network for community discovery is an emerging research area because current methods do not focus on social graphs or they are not efficient for large-scale datasets.…”

“…In the work of Li and Schuurmans, the authors introduced an iterative borough strategy to identify communities that is coupled with a fast constrained power method that sequentially achieves tighter spectral relaxations. Some fuzzy community detection algorithms calculate the possibility of each node belonging to every community, such as in other works . The authors of these approaches provide a fresh idea for the understanding of community detection.…”

Review of Community Detection over Social Media:Graph Prospective

“…Yafang Li [29] work over rank-based community structure grouping web pages through page pank centrality algorithm. Samira Malek et al [30] work over fuzzy based duo centric overlapped community detection. YunfengXu et al [31] work over biological structure to analysis strength and backbone degree of social network for member-based community detection.…”

MOFSocialNet: Exploiting Metal-Organic Framework Relationships via Social Network Analysis

“…One other parameter, which we looked at in MOFSocialNet, is community detection [28][29][30][31][32][33][34]. Community detection is essentially a type of clustering problem.…”