Overlapping Community Detection Extended from Disjoint Community Structure

Abstract

Community detection is a hot issue in the study of complex networks. Many community detection algorithms have been put forward in different fields. But most of the existing community detection algorithms are used to find disjoint community structure. In order to make full use of the disjoint community detection algorithms to adapt to the new demand of overlapping community detection, this paper proposes an overlapping community detection algorithm extended from disjoint community structure by selecting overlapping nodes (ONS-OCD). In the algorithm, disjoint community structure with high qualities is firstly taken as input, then, potential members of each community are identified. Overlapping nodes are determined according to the node contribution to the community. Finally, adding

Research on Deep Integration of the “Belt and Road” Aviation Network Based on Community Structure from the Perspective of China

“…Since the edge betweenness-based GN algorithm was designed [12], numerous community detection algorithms have been proposed successively to solve the community detection issues in various sectors. Currently available community detection algorithms are mostly designed for static networks, including 5 types, namely modularity optimization, tag propagation, local expansion, streaming analysis, and deep learning [13]. Each of the 5 types of algorithms has its own features.…”

CILPA: An Overlapping Community Detection Algorithm Based on Label Propagation and Node Information

“…Scholars have proposed different approaches and adopted different perspectives to detect overlapping communities, including fuzzy clustering, local extension, density peaking, faction filtering, and label propagation methods. Palla et al [12] proposed the existence of overlapping communities in complex networks in 2005, marking the first theoretical study of "overlapping communities" in complex systems. Specifically, the NDOCD algorithm [5] they proposed removes links in a network based on the largest subgroup, and then clusters the nodes to discern the community structure.…”

“…To redress the overlapping communities detection for weighted networks, Farkas et al [6] proposed the faction-based CPMw algorithm, which filters k-clusters with complete subgraphs smaller than a threshold and retains clusters in the network larger than that threshold. The algorithm then performs the same merge extension operation to discover communities as in literature [12].…”

“…The LFM algorithm [11] selects nodes as seed communities, and then extends and merges neighbouring nodes to obtain the optimal fitness. Xing et al [12] posited that overlapping community structures can be discovered by referencing disjoint communities. Ni et al [13] set out a framework for local overlapping community detection that can effectively identify community structures.…”

“…In recent years, research on overlapping community discovery has made great progress. Developments in this area include: 1) Faction filtering based on complete subgraphs to obtain overlapping communities [5,6]; 2) Fuzzy clustering [7][8][9][10], local extension [11][12][13][14][15], density peaking [16][17][18] and label propagation [19][20][21][22][23][24] based on node to discover overlapping clusters; 3) Graph partitioning based on edge to obtain overlapping communities [25][26][27]; 4) Node representation-based learning to identify overlapping communities [24,28]. Due to the complexity and diversity of real-world large-scale networks, improving the processing performance and accuracy of overlapping community discovery algorithms has been intensely researched.…”